WO2021212274A1 - 疲劳驾驶状态检测方法、装置、计算机设备和存储介质 - Google Patents

疲劳驾驶状态检测方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2021212274A1
WO2021212274A1 PCT/CN2020/085627 CN2020085627W WO2021212274A1 WO 2021212274 A1 WO2021212274 A1 WO 2021212274A1 CN 2020085627 W CN2020085627 W CN 2020085627W WO 2021212274 A1 WO2021212274 A1 WO 2021212274A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
judgment threshold
steering wheel
threshold
driving
Prior art date
Application number
PCT/CN2020/085627
Other languages
English (en)
French (fr)
Inventor
于琪
Original Assignee
南京天擎汽车电子有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京天擎汽车电子有限公司 filed Critical 南京天擎汽车电子有限公司
Priority to PCT/CN2020/085627 priority Critical patent/WO2021212274A1/zh
Publication of WO2021212274A1 publication Critical patent/WO2021212274A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers

Definitions

  • the present application provides a method, a device, a computer device, and a storage medium for detecting a fatigue driving state.
  • the fatigue driving detection technology mainly includes the following methods:
  • the detected physiological signals mainly include EEG, ECG, EMG, etc.
  • the detection method based on image vision technology mainly collects the driver's facial signal through the camera, and judges whether the driver is fatigued according to the driver's eye state or head tilt angle and rotation angle.
  • a detection method based on the characteristics of the driver's operating behavior that is, judging whether the driver is fatigued through vehicle driving information such as vehicle speed, lateral position of the lane, yaw angle, steering wheel angle, accelerator pedal, and brake pedal.
  • the above method (1) requires a special instrument to be worn on the driver to monitor the signal, and the operation is more complicated.
  • the above method (2) requires complex image processing, the operation is more complicated, and the amount of data processing is large, and the efficiency is low.
  • the above methods (3) and (4) are the detection methods based on the driver's operating behavior characteristics. Due to the convenience of real-time data collection, low cost, easy popularization, and no interference to the driver in the process of driving the vehicle, they become the detection method of fatigue driving. Important research direction. However, the traditional detection methods based on driver's operating behavior characteristics are generally based on one feature, which is too single and prone to false alarms; or data training is carried out through SVM (support vector machine). Multiple feature information is considered comprehensively, but the number of training samples needs to be large enough. Once the actual vehicle calibration parameters are fixed, they cannot be modified, and the individual differences cannot be adjusted adaptively, so the accuracy is low.
  • a method, a device, a computer device, and a storage medium for detecting a fatigue driving state are provided.
  • a method for detecting fatigue driving state including:
  • the driver driving the vehicle is in a fatigue driving state.
  • a fatigue driving state detection device including:
  • Duration and speed acquisition module used to obtain the current driving time and current driving speed of the vehicle
  • the reference parameter value collection module is configured to obtain the reference parameter value of the vehicle according to the predetermined reference parameter when the current driving time length and the current driving speed meet the preset requirements;
  • a judgment threshold query module configured to query a pre-configured judgment threshold corresponding to the reference parameter
  • the first judgment module is used to judge whether the reference parameter value meets the judgment threshold.
  • the judgment module is configured to, when the reference parameter value meets the judgment condition based on the judgment threshold, the driver driving the vehicle is in a fatigue driving state.
  • a computer device including a memory and one or more processors, the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the one or more processors execute The following steps:
  • the driver driving the vehicle is in a fatigue driving state.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions.
  • the computer-readable instructions When executed by one or more processors, the one or more processors perform the following steps:
  • the driver driving the vehicle is in a fatigue driving state.
  • Fig. 1 is an application environment diagram of the fatigue state detection method according to one or more embodiments.
  • Fig. 2 is a schematic flowchart of a fatigue state detection method according to one or more embodiments.
  • Fig. 3 is a flowchart of a configuration method of a judgment threshold according to one or more embodiments.
  • Fig. 4 is a schematic flowchart of a fatigue state detection method according to another or more embodiments.
  • Fig. 5 is a block diagram of a fatigue state detection device according to one or more embodiments.
  • Figure 6 is a block diagram of a computer device according to one or more embodiments.
  • the fatigue driving state detection method provided in this application can be applied to the application environment as shown in FIG. 1.
  • the vehicle terminal 102 communicates with each controller 104 on the vehicle to collect the current driving time, current driving speed, and reference parameter values of the vehicle.
  • the vehicle terminal 102 determines that the current driving time and current driving speed of the vehicle meet the preset requirements, According to the predetermined reference parameter, the reference parameter value of the vehicle is obtained, and then the judgment threshold corresponding to the reference parameter configured during normal driving is queried, so that it can be judged whether the reference parameter value meets the judgment threshold, and if so, it can be judged to drive the vehicle Of drivers are in a state of fatigue driving.
  • the above fatigue driving state detection method introduces a custom judgment threshold, that is, each driver sets the judgment threshold corresponding to the reference parameter configured during normal driving during use, so that individual differences are fully considered, and the judgment threshold is guaranteed Accuracy, which can improve the accuracy of the fatigue driving state detection, thereby reducing the false alarm rate.
  • the vehicle terminal 102 may be, but is not limited to, various intelligent control devices installed on the vehicle, including a personal computer, a notebook computer, a smart phone, and a tablet computer.
  • the controller is a variety of controllers installed on the vehicle to collect vehicle information.
  • a method for detecting a fatigue driving state is provided. Taking the method applied to the vehicle terminal in FIG. 1 as an example for description, the method includes the following steps:
  • S202 Acquire the current driving time and current driving speed of the vehicle.
  • the current running time of the vehicle may be the length of time the vehicle power is turned on as recorded by the vehicle terminal, and the current running speed of the vehicle is collected by the vehicle terminal through the speed sensor.
  • the vehicle terminal can collect the current driving time and current speed of the vehicle in real time after the vehicle is running.
  • acquiring the current driving time and current driving speed of the vehicle includes: acquiring the power state and the current driving speed of the vehicle; and counting the current driving time of the vehicle when the power state is in the on state.
  • the power state is taken into account, and the accuracy of the driving time statistics is improved. This fully takes into account that the fatigue driving risk index is relatively low when the vehicle speed is low, and there may be false alarms of fatigue driving, thereby increasing fatigue. Accuracy of driving status detection.
  • the vehicle terminal may include the following three situations when counting the driving time: First, the power state is ON, the vehicle speed is valid, and the vehicle speed is greater than or equal to a preset speed, for example, when the driving time is 5km/h. Second, when the power state changes to ACC or OFF gear, the driving time is cleared, and the time is restarted when the conditions are met again. Third, the power state is ON, the vehicle speed is invalid or the vehicle speed is less than the preset speed, for example, 5km/h exceeds the vehicle stop timing duration, that is, the driving duration is cleared when the VehicleStopTimer is satisfied, and the timing is restarted when the conditions are met again.
  • the preset requirements refer to the minimum requirements for determining the fatigue driving state.
  • the vehicle speed can be limited to 30km/h or more, in other embodiments it can be limited to other speeds, the current driving time can be limited to half an hour, etc., in other embodiments it can be limited to other time lengths.
  • the pre-determined reference parameters are obtained by collecting driver fatigue driving data, analyzing the collected data, and extracting parameters related to fatigue driving.
  • the reference parameters include, but are not limited to, the proportion of the total operating time of the accelerator brake per unit time, and the maximum steering wheel deviation.
  • the above-mentioned reference parameters can also be replaced with other effective feature reference parameters related to fatigue driving.
  • basic data such as vehicle speed, steering wheel angle, accelerator pedal, brake pedal, and yaw rate can be collected and statistically analyzed to obtain reference parameter values.
  • a formula for calculating the reference parameter value may be pre-stored in the vehicle terminal, so that after the vehicle terminal collects the above-mentioned basic data, the corresponding reference parameter value is obtained according to the formula.
  • the judgment threshold is the judgment threshold corresponding to the reference parameter configured by the driver when the vehicle is running normally before or during the use of the vehicle. For example, when the driver uses the vehicle for the first time, it can The corresponding configuration switch to turn on the configuration state, and configure. After the driver has used it many times, he can turn on the configuration state through the corresponding configuration switch on the control screen to update the original configuration. For example, after a period of normal driving, after N seconds of data are counted, use this segment of data to update the judgment Threshold, and store the configured judgment threshold in association with the corresponding driver’s profile picture.
  • the vehicle terminal may query the pre-configured judgment threshold corresponding to the reference parameter when the current driving time and the current driving speed meet the preset requirements, and the query step and the above-mentioned step of obtaining the reference parameter value may be parallel.
  • the vehicle terminal may first collect the driver's avatar, and then match the collected avatar with the pre-stored avatar, and obtain the judgment threshold corresponding to the avatar with the greatest matching degree as the threshold used for this judgment.
  • judging whether the reference parameter value meets the judging threshold can be performed by the vehicle terminal according to a preset judgment logic. For example, the proportion of the total operating time of the characteristic index accelerator brake in a unit time is less than the second threshold P th , and the steering wheel is the largest in a unit time.
  • the immobile time is greater than the third threshold T ⁇ _th , and the total number of accelerators and brakes per unit time is less than the fourth threshold Freq th , it is determined that the driver of the driving vehicle is in a fatigued driving state, as long as one is not satisfied, the driver of the driving vehicle is determined Not in a state of fatigue driving.
  • the above fatigue driving state detection method introduces a custom judgment threshold, that is, each driver sets the judgment threshold corresponding to the reference parameter configured during normal driving during use, so that individual differences are fully considered, and the judgment threshold is guaranteed Accuracy, which can improve the accuracy of the fatigue driving state detection, thereby reducing the false alarm rate.
  • the detection will only be performed when the current driving time and current driving speed meet the preset requirements, which fully takes into account the fatigue driving at low speeds
  • the risk index is relatively low, and there may be false alarms of fatigue driving, which in turn improves the accuracy of fatigue driving state detection.
  • the configuration method of determining the threshold includes: receiving a threshold configuration instruction input by the driver; collecting the driving parameter value of the vehicle when the vehicle is traveling according to the threshold configuration instruction, and the driving parameter value corresponds to the reference parameter; and calculating according to the driving parameter value Determine the threshold and configure it.
  • the specific judgment threshold configuration includes two stages: the first stage is when the driver drives the vehicle for the first time, the configuration is mandatory, and the second stage is when the driver has used it for a period of time, the first stage or the last setting needs to be updated Judgment threshold.
  • the driver can open the configuration of the judgment threshold by clicking the judgment threshold configuration button on the control screen.
  • the vehicle terminal in the second stage can make judgments, such as judging whether the use time of the judgment threshold is greater than the preset time, for example If it is more than half a year or a year, it is prompted to update the judgment threshold, so that the driver can start the configuration of the judgment threshold by clicking the judgment threshold configuration button on the control screen.
  • the configuration method of the judgment threshold may be that the vehicle terminal receives the threshold configuration instruction input by the driver, and then the vehicle terminal opens the configuration state, so that the vehicle terminal collects the vehicle speed, steering wheel angle, accelerator pedal, brake pedal and Basic data such as yaw rate, then carry out statistical analysis to obtain the driving parameter value corresponding to the reference parameter, and finally calculate and configure the judgment threshold according to the driving parameter value, for example, take the maximum value of the driving parameter value as the judgment threshold.
  • the method for obtaining the judgment threshold corresponding to the maximum immobilization time of the steering wheel includes: obtaining the steering wheel angle of each time the steering wheel does not move within a preset time period; The maximum difference is obtained as the steering wheel immobility threshold; the difference of the steering wheel angle in the statistical unit time window is less than the first time difference of the steering wheel immobility threshold; the largest first time difference is regarded as the maximum steering wheel immobility time corresponding Judgment threshold.
  • the user can customize the steering wheel angle immobilization threshold, and the threshold ⁇ is adopted by default.
  • the specific settings can be implemented in the form of custom setting threshold switches on the large screen. After counting N seconds of data to obtain the threshold, close the setting function.
  • the steering wheel immobilization threshold is first determined, so that the difference of the steering wheel angle in the unit time window is calculated and the first time difference is less than the steering wheel immobilization threshold. Can guarantee accuracy.
  • the configuration of the judgment threshold corresponding to the total number of accelerator brakes includes: collecting the frequency of the accelerator opening change in a unit time window as the accelerator pedal operation frequency; collecting the frequency of the brake pedal change in the unit time window , As the operating frequency of the brake pedal; get the total number of accelerator brakes based on the operating frequency of the accelerator pedal and the operating frequency of the brake pedal.
  • the configuration of the judgment threshold corresponding to the proportion of the total operation time of the accelerator brake in a unit time includes: collecting the time when the accelerator opening is non-zero in the unit time window as the accelerator operation time; collecting the brake in the unit time window The time of the non-zero point of the cylinder pressure is used as the brake pedal operation time; according to the accelerator operation time, the brake pedal operation time and the time length of the unit time window, the proportion of the total accelerator brake operation time per unit time is calculated.
  • the driver can customize the judgment threshold setting, and fully consider the performance of the vehicle when setting, so that the judgment threshold is more in line with the driver and the current vehicle, and the accuracy of the judgment threshold is ensured.
  • the method before receiving the threshold configuration instruction input by the driver, the method further includes: collecting the driver's avatar through the image acquisition device; judging whether there is a judgment threshold corresponding to the driver according to the avatar; When the judgment threshold is corresponding to the member, the judgment threshold configuration prompt will be output.
  • the image acquisition device can be a device that can take a picture of the driver’s portrait.
  • the vehicle terminal can recognize the driver’s avatar sitting in the driver’s seat by position.
  • the avatar must include at least the driver’s face.
  • the judgment threshold is stored in association with the driver's avatar, so the corresponding judgment threshold can be queried based on the face.
  • Threshold configuration prompt it is determined in advance whether the vehicle has a judgment threshold corresponding to the driver based on the driver’s avatar collected by the image acquisition device. In this way, when there is no storage, the driver can be prompted to configure and ensure the personality of the judgment threshold. Therefore, the accuracy of the detection of the fatigue driving state can be improved, and the false alarm rate can be reduced.
  • the judgment threshold configuration prompt is output.
  • the generation time refers to the storage time of the judgment threshold, that is, the time when the judgment threshold is stored in the vehicle terminal, so that the vehicle terminal can obtain the use time of the judgment threshold according to the current system time and the generation time.
  • the preset duration is set by the user in advance, it can be half a year or a year, etc.
  • the usage time of the judgment threshold is greater than the preset time, it means that the usage time of the judgment threshold is longer, which may be caused by too long time. There is an error in the judgment threshold, so the vehicle terminal prompts the driver to reconfigure.
  • it is determined in advance whether the vehicle has a judgment threshold corresponding to the driver based on the driver’s profile picture collected by the image acquisition device. In this way, when the judgment threshold is stored, it is also necessary to obtain the usage time of the judgment threshold. The problem of error in the judgment threshold caused by too long time is avoided, and the personalization of the judgment threshold is ensured, thereby improving the accuracy of detecting the fatigue driving state, thereby reducing the false alarm rate.
  • the configuration method of the judgment threshold mainly includes: the vehicle terminal receives the threshold input by the driver Configure the instruction, and then the vehicle terminal collects the driver’s avatar through the image acquisition device, and determines whether the vehicle terminal stores a judgment threshold corresponding to the collected avatar, so that when it does not exist, it outputs a judgment threshold configuration prompt, that is to say While outputting the judgment threshold configuration prompt, the vehicle terminal collects the driving parameter value of the vehicle during driving according to the threshold configuration instruction, and the driving parameter value corresponds to the reference parameter; the judgment threshold is calculated and configured according to the driving parameter value.
  • the vehicle terminal obtains the generation time of the judgment threshold; obtains the use duration of the judgment threshold according to the current system time and generation time; determines whether the use duration is greater than the preset duration; when the use duration is greater than the preset duration
  • the judgment threshold configuration prompt is output, that is, the driving parameter value of the vehicle during driving is collected according to the threshold configuration instruction, and the driving parameter value corresponds to the reference parameter; the judgment threshold is calculated and configured according to the driving parameter value. If there is a judgment threshold and the use time of the judgment threshold is less than or equal to the preset time period, a prompt that the judgment threshold already exists can be directly output, and the user can choose whether to configure it again.
  • the reference parameters include, but are not limited to, the proportion of the total operating time of the accelerator brake in a unit time, the maximum immobilization time of the steering wheel, and the total number of times of accelerator brake.
  • a plurality of driver behavior characteristics are comprehensively considered, and the fatigue driving state can be judged in various aspects, and the detection accuracy of the fatigue driving state can be improved.
  • the method of collecting the maximum immobilization time of the steering wheel includes: counting the second time difference in which the steering wheel angle difference in the unit time window is less than the steering wheel immobilization threshold; and using the largest second time difference as the maximum immobilization time of the steering wheel .
  • the vehicle terminal detects that the steering wheel angle difference between adjacent times is less than the threshold ⁇ , it is considered that the steering wheel is not moving. In this way, the vehicle terminal counts the continuous immobilization time of the steering wheel in the unit time window T, and the maximum value is taken as the maximum immobilization time T ⁇ of the steering wheel.
  • the second time difference of the steering wheel angle difference in the unit time window is less than the steering wheel immobilization threshold to obtain the maximum immobilization time of the steering wheel in real time to improve accuracy.
  • the method of collecting the total number of accelerator brakes includes: collecting the frequency of the accelerator opening change in the unit time window as the accelerator pedal operation frequency; collecting the frequency of the brake pedal change in the unit time window as the brake pedal Operating frequency: Get the total number of throttle brakes based on the operating frequency of the accelerator pedal and the operating frequency of the brake pedal.
  • the total number of throttle and brake operations per unit time, Freq is the sum of throttle operation frequency and brake operation frequency per unit time:
  • the frequency of the accelerator pedal operation is expressed by the frequency of the accelerator opening change
  • the frequency of the brake pedal operation is expressed by the frequency of the brake pedal change, which is convenient for detection and has high accuracy.
  • the method of collecting the proportion of the total operating time of the accelerator brake in a unit time includes: collecting the non-zero time of the accelerator opening in the unit time window as the accelerator operation time; collecting the non-zero value of the brake cylinder pressure in the unit time window The time at zero is regarded as the operating time of the brake pedal; according to the accelerator operating time, the operating time of the brake pedal and the time length of the unit time window, the proportion of the total operating time of the accelerator brake in the unit time is calculated.
  • Throttle operating time Trt time is calculated by the throttle opening non-zero point:
  • the brake pedal operation time Brk time is expressed by the number of non-zero points of the brake cylinder pressure:
  • the accelerator operation time is expressed by the time when the accelerator opening is non-zero
  • the brake pedal operation time is expressed by the time when the brake cylinder pressure is non-zero, which is convenient for detection and has high accuracy.
  • judging whether the reference parameter value meets the judging threshold includes: judging whether the maximum immobilization time of the steering wheel is greater than the first judging threshold, whether the proportion of the total operating time of the throttle and braking in unit time is less than the second judging threshold, and the total throttle and braking time Whether the number of times is less than the third judgment threshold; when the maximum immobilization time of the steering wheel is greater than the first judgment threshold, the total operation time of throttle and brake in unit time is less than the second judgment threshold, and the total number of throttle brakes is less than the third judgment threshold, then refer to the parameter The value meets the judgment threshold.
  • a plurality of driver behavior characteristics are comprehensively considered, and the fatigue driving state can be judged in various aspects, and the detection accuracy of the fatigue driving state can be improved.
  • FIG. 4 is a flowchart of a method for detecting a fatigue driving state in another embodiment.
  • the vehicle terminal first collects the driving time and vehicle speed, and if the driving time and vehicle speed meet the preset conditions For example, when the driving time exceeds T 0 and the vehicle speed is greater than 30km/h, it is necessary to confirm whether the driver is in a fatigued driving state, so that the vehicle terminal collects basic data, such as steering wheel angle, accelerator and brake data, etc., and then the vehicle terminal calculates each reference
  • the parameter value is the detection of fatigue driving through the three characteristics of integrated steering wheel immobility time, accelerator brake pedal stepping frequency and the proportion of the total accelerator brake operation time.
  • Vehicle terminal may set a time window T, the repetition rate of two successive time windows Tr, that is to say each time statistical characteristics T a fatigue driving state obtained before a time T after time T and a repetition time T R < equivalent Every time (TT r ), output the maximum immobility time T ⁇ of the steering wheel within the statistical unit time T of the fatigue driving state, the total time of the throttle brake operation accounted for P, the total number of throttle brake operations Freq, and then the vehicle terminal will make the following judgments in turn: The total time of throttle and brake operation P ⁇ P th ; the total number of throttle and brake operations Freq ⁇ Freq th ; the maximum immobility time of the steering wheel T ⁇ > T ⁇ _th ; if all the above conditions are met, the driver is considered to be in a state of fatigue. Otherwise, the driver is not in a state of fatigue driving.
  • the above fatigue driving state detection method introduces a custom judgment threshold, that is, each driver sets the judgment threshold corresponding to the reference parameter configured during normal driving during use, so that individual differences are fully considered, and the judgment threshold is guaranteed Accuracy, which can improve the accuracy of the fatigue driving state detection, thereby reducing the false alarm rate.
  • the detection will only be performed when the current driving time and current driving speed meet the preset requirements, which fully takes into account the fatigue driving at low speeds
  • the risk index is relatively low, and there may be false alarms of fatigue driving, which in turn improves the accuracy of fatigue driving state detection.
  • a device for detecting a fatigue driving state including: a duration and speed collection module 100, a reference parameter value collection module 200, a judgment threshold query module 300, a first judgment module 400, and The determination module 500, wherein:
  • the duration and speed acquisition module 100 is used to acquire the current travel duration and current travel speed of the vehicle
  • the reference parameter value collection module 200 is configured to obtain the reference parameter value of the vehicle according to the predetermined reference parameter when the current driving time and the current driving speed meet the preset requirements;
  • the judgment threshold query module 300 is used to query the pre-configured judgment threshold corresponding to the reference parameter
  • the first judgment module 400 is used to judge whether the reference parameter value meets the judgment threshold
  • the judging module 500 is used for when the reference parameter value meets the judging condition based on the judging threshold, the driver driving the vehicle is in a fatigued driving state.
  • the above-mentioned fatigue driving state detection device may further include:
  • the receiving module is used to receive the threshold configuration instruction input by the driver
  • the parameter collection module is used to collect the driving parameter value when the vehicle is driving according to the threshold configuration instruction, and the driving parameter value corresponds to the reference parameter;
  • the judgment threshold configuration module is used to calculate and configure the judgment threshold according to the value of the driving parameter.
  • the above-mentioned fatigue driving state detection device may further include:
  • the image acquisition module is used to collect the driver's avatar through the image acquisition device
  • the second judgment module is used to judge whether the judgment threshold corresponding to the driver is stored according to the avatar;
  • the first prompt module is used to output a judgment threshold configuration prompt when the judgment threshold corresponding to the driver is not stored.
  • the above-mentioned fatigue driving state detection device may further include:
  • the generation time acquisition module is used to acquire the generation time of the judgment threshold when the judgment threshold corresponding to the driver is stored;
  • the usage duration acquisition module is used to obtain the usage duration of the judgment threshold value according to the current system time and generation time;
  • the third judgment module is used to judge whether the use time is greater than the preset time
  • the second prompt module is used to output a judgment threshold configuration prompt when the use time is greater than the preset time.
  • the reference parameters include, but are not limited to, the proportion of the total operating time of the accelerator brake in a unit time, the maximum immobilization time of the steering wheel, and the total number of times of accelerator brake.
  • the above-mentioned fatigue driving state detection device may further include:
  • the steering wheel angle acquisition module is used to acquire the steering wheel angle every time the steering wheel does not move in a preset time period
  • the steering wheel immobility threshold acquisition module is used to calculate the difference between the steering wheel angles when the steering wheel is not moving two adjacent times, and obtain the maximum difference as the steering wheel immobility threshold;
  • the first time difference statistics module is used to count the first time difference when the steering wheel angle difference in the unit time window is less than the steering wheel immobility threshold;
  • the first configuration module is configured to use the largest first time difference as the judgment threshold corresponding to the maximum immobilization time of the steering wheel.
  • the above-mentioned fatigue driving state detection device may further include:
  • the second time difference statistics module is used to count the second time difference when the steering wheel angle difference in the unit time window is less than the steering wheel immobility threshold;
  • the second configuration module is used to use the largest second time difference as the maximum stationary time of the steering wheel.
  • the above-mentioned fatigue driving state detection device may further include:
  • the accelerator pedal operation frequency acquisition module is used to collect the frequency of the accelerator opening change in the unit time window as the accelerator pedal operation frequency
  • the brake pedal operating frequency collection module is used to collect the frequency of the brake pedal change in a unit time window as the brake pedal operating frequency
  • the calculation module of the total number of throttle brakes is used to obtain the total number of throttle brakes according to the operating frequency of the accelerator pedal and the operating frequency of the brake pedal.
  • the above-mentioned fatigue driving state detection device may further include:
  • Throttle operation time acquisition module used to collect the time when the throttle opening is not zero in the unit time window as the throttle operation time
  • the brake pedal operation time collection module is used to collect the non-zero point of the brake cylinder pressure in the unit time window as the brake pedal operation time;
  • the proportion statistics module is used to calculate the proportion of the total accelerator brake operation time per unit time according to the accelerator operation time, the brake pedal operation time and the time length of the unit time window.
  • the above-mentioned first judgment module 400 may include:
  • the output unit is used for when the maximum immobilization time of the steering wheel is greater than the first judgment threshold, the proportion of the total operation time of the throttle brake in unit time is less than the second judgment threshold, and the total number of throttle brakes is less than the third judgment threshold, then the reference parameter value meets the judgment Threshold.
  • the aforementioned duration and speed collection module 100 may include:
  • the first acquiring unit is used to acquire the power state and the current driving speed of the vehicle
  • the second acquiring unit is used to count the current driving time of the vehicle when the power state is in the on state.
  • each module in the above-mentioned fatigue driving state detection device can be implemented in whole or in part by software, hardware and a combination thereof.
  • the above-mentioned modules may be embedded in the form of hardware or independent of the processor in the computer equipment, or may be stored in the memory of the computer equipment in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.
  • a computer device is provided.
  • the computer device may be a terminal, and its internal structure diagram may be as shown in FIG. 6.
  • the computer equipment includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus.
  • the processor of the computer device is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system and computer readable instructions.
  • the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen
  • the input device of the computer equipment can be a touch layer covered on the display screen, or it can be a button, trackball or touch pad set on the housing of the computer equipment , It can also be an external keyboard, touchpad, or mouse.
  • FIG. 6 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied.
  • the specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.
  • a computer device includes a memory and one or more processors, and the memory stores computer-readable instructions.
  • the one or more processors perform the following steps: Obtain the current driving of a vehicle Duration and current driving speed; when the current driving time and current driving speed meet the preset requirements, obtain the reference parameter value of the vehicle according to the predetermined reference parameter; query the pre-configured judgment threshold corresponding to the reference parameter; judge the reference parameter Whether the value meets the judgment threshold; when the reference parameter value meets the judgment condition based on the judgment threshold, the driver driving the vehicle is in a state of fatigue driving.
  • the configuration method of the judgment threshold involved when the processor executes the computer-readable instruction includes: receiving a threshold configuration instruction input by the driver; collecting the driving parameter value of the vehicle when the vehicle is driving according to the threshold configuration instruction, and the driving parameter value is Corresponding to reference parameters; the judgment threshold is calculated and configured according to the value of the driving parameter.
  • the processor before receiving the threshold configuration instruction input by the driver when the processor executes the computer-readable instruction, it further includes: collecting the driver's avatar through the image acquisition device; When the judgment threshold corresponding to the driver is not stored, the judgment threshold configuration prompt will be output.
  • the processor executes the computer-readable instruction to determine whether the judgment threshold corresponding to the driver is stored according to the avatar, it further includes: when the judgment threshold corresponding to the driver is stored, obtaining the judgment The generation time of the threshold; the use time of the judgment threshold is obtained according to the current system time and the generation time; it is judged whether the use time is greater than the preset time; when the use time is greater than the preset time, the judgment threshold configuration prompt is output.
  • the reference parameters involved when the processor executes the computer-readable instructions include, but are not limited to, the proportion of the total operating time of the accelerator brake in a unit time, the maximum immobilization time of the steering wheel, and the total number of times of accelerator brake.
  • the method for obtaining the judgment threshold corresponding to the maximum immobilization time of the steering wheel involved when the processor executes the computer-readable instruction includes: obtaining the steering wheel angle of each time the steering wheel does not move within a preset time period; and calculating adjacent The difference between the two steering wheel angles when the steering wheel is not moving, and the largest difference is obtained as the steering wheel immobility threshold; the difference of the steering wheel angle in the statistical unit time window is less than the first time difference of the steering wheel immobility threshold; the largest A time difference is used as the judgment threshold corresponding to the maximum immobilization time of the steering wheel.
  • the method of collecting the maximum immobilization time of the steering wheel involved when the processor executes the computer-readable instruction includes: counting the second time difference in which the steering wheel angle difference in the unit time window is less than the steering wheel immobilization threshold; The second time difference is regarded as the maximum immobility time of the steering wheel.
  • the method of collecting the total number of accelerator brakes involved when the processor executes the computer-readable instruction includes: collecting the frequency of the accelerator opening change in a unit time window as the accelerator pedal operating frequency; and collecting the frequency of the accelerator pedal operation in the unit time window.
  • the frequency at which the brake pedal changes is used as the operating frequency of the brake pedal; the total number of times of accelerator braking is obtained according to the operating frequency of the accelerator pedal and the operating frequency of the brake pedal.
  • the method of collecting the proportion of the total operating time of the accelerator brake in the unit time involved when the processor executes the computer-readable instruction includes: collecting the time when the accelerator opening is non-zero in the unit time window as the accelerator operating time; collecting The time when the brake cylinder pressure is non-zero in the unit time window is used as the brake pedal operation time; according to the accelerator operation time, the brake pedal operation time and the time length of the unit time window, the proportion of the total accelerator brake operation time per unit time is calculated.
  • the judgment of whether the reference parameter value meets the judgment threshold realized by the processor when the processor executes the computer-readable instruction includes: judging whether the maximum immobilization time of the steering wheel is greater than the first judgment threshold, and the percentage of the total accelerator brake operation time per unit time Whether it is less than the second judgment threshold, whether the total number of throttle brakes is less than the third judgment threshold; when the maximum immobilization time of the steering wheel is greater than the first judgment threshold, the total operating time of the throttle brake in unit time is less than the second judgment threshold, and the total number of throttle brakes When it is less than the third judgment threshold, the reference parameter value meets the judgment threshold.
  • acquiring the current driving time and current driving speed of the vehicle when the processor executes the computer-readable instructions includes: acquiring the power state and the current driving speed of the vehicle; and counting the current driving time of the vehicle when the power state is on. Driving time.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions.
  • the one or more processors perform the following steps: Obtain the current Driving time and current driving speed; when the current driving time and current driving speed meet the preset requirements, obtain the reference parameter value of the vehicle according to the predetermined reference parameter; query the pre-configured judgment threshold corresponding to the reference parameter; judgment reference Whether the parameter value meets the judgment threshold; when the reference parameter value meets the judgment condition based on the judgment threshold, the driver driving the vehicle is in a state of fatigue driving.
  • the configuration of the judgment threshold involved when the computer-readable instruction is executed by the processor includes: receiving a threshold configuration instruction input by the driver; collecting the driving parameter value when the vehicle is driving according to the threshold configuration instruction, and the driving parameter value Corresponding to the reference parameter; the judgment threshold is calculated and configured according to the value of the driving parameter.
  • the method before receiving the threshold configuration instruction input by the driver realized when the computer-readable instruction is executed by the processor, the method further includes: collecting the driver's avatar through the image acquisition device; The corresponding judgment threshold; when the judgment threshold corresponding to the driver is not stored, the judgment threshold configuration prompt is output.
  • the computer-readable instruction after the computer-readable instruction is executed by the processor to determine whether a judgment threshold corresponding to the driver is stored according to the avatar, it further includes: when the judgment threshold corresponding to the driver is stored, obtaining Determine the generation time of the threshold; obtain the use time of the judgment threshold according to the current system time and generation time; determine whether the use time is greater than the preset time; when the use time is greater than the preset time, output the judgment threshold configuration prompt.
  • the reference parameters involved when the computer-readable instructions are executed by the processor include, but are not limited to, the proportion of the total operating time of the accelerator brake in a unit time, the maximum stationary time of the steering wheel, and the total number of times of accelerator brake.
  • the method for obtaining the judgment threshold corresponding to the maximum immobilization time of the steering wheel involved when the computer-readable instruction is executed by the processor includes: obtaining the steering wheel angle each time the steering wheel is immobile in a preset time period; and calculating the phase The difference between the steering wheel angles when the steering wheel is not moving twice, and the largest difference is obtained as the steering wheel immobility threshold; the difference of the steering wheel angle in the statistical unit time window is less than the first time difference of the steering wheel immobility threshold; the largest The first time difference is used as the judgment threshold corresponding to the maximum immobilization time of the steering wheel.
  • the method of collecting the maximum immobilization time of the steering wheel involved when the computer-readable instruction is executed by the processor includes: a second time difference in which the steering wheel angle difference in a statistical unit time window is less than the steering wheel immobility threshold; The largest second time difference is used as the maximum stationary time of the steering wheel.
  • the method of collecting the total number of accelerator brakes involved when the computer-readable instruction is executed by the processor includes: collecting the frequency of the accelerator opening change in a unit time window as the accelerator pedal operating frequency; collecting the unit time window The frequency at which the internal brake pedal changes is used as the operating frequency of the brake pedal; the total number of times of accelerator braking is obtained according to the operating frequency of the accelerator pedal and the operating frequency of the brake pedal.
  • the method of collecting the proportion of the total throttle brake operation time per unit time involved when the computer-readable instruction is executed by the processor includes: collecting the non-zero throttle opening time in the unit time window as the throttle operation time; Collect the non-zero point of the brake cylinder pressure in the unit time window as the brake pedal operation time; calculate the proportion of the total accelerator brake operation time per unit time based on the accelerator operation time, the brake pedal operation time and the time length of the unit time window.
  • the judgment of whether the reference parameter value meets the judgment threshold realized when the computer-readable instruction is executed by the processor includes: judging whether the maximum immobilization time of the steering wheel is greater than the first judgment threshold, and the total accelerator brake operation time per unit time accounts for Whether the ratio is less than the second judgment threshold and whether the total number of throttle brakes is less than the third judgment threshold; when the maximum immobilization time of the steering wheel is greater than the first judgment threshold, the total operating time of the accelerator brake in unit time is less than the second judgment threshold, and the total throttle brake operation When the number of times is less than the third judgment threshold, the reference parameter value meets the judgment threshold.
  • acquiring the current driving time and current driving speed of the vehicle when the computer-readable instruction is executed by the processor includes: acquiring the power state and the current driving speed of the vehicle; The current driving time.
  • Non-volatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Channel (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Mathematical Physics (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种疲劳驾驶状态检测方法，包括：获取车辆的当前行驶时长和当前行驶速度；当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；查询预先配置的与所述参考参数对应的判断阈值；判断所述参考参数值是否符合所述判断阈值；当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。

Description

疲劳驾驶状态检测方法、装置、计算机设备和存储介质 技术领域
本申请一种疲劳驾驶状态检测方法、装置、计算机设备和存储介质。
背景技术
随着人工智能技术的发展，出现了疲劳驾驶检测技术，可以检测驾驶行驶中的车辆的人是否处于疲劳驾驶状态。该疲劳驾驶检测技术主要包括以下几种方法：
(1)基于驾驶人生理信号的检测方法，检测的生理信号主要包括脑电信号EEG、心电信号ECG、肌电信号EMG等。
(2)基于图像视觉技术的检测方法，主要是通过摄像头采集驾驶员面部信号，根据驾驶员眼睛状态或头部倾斜角度及旋转角度判断驾驶员是否疲劳。
(3)基于驾驶员操作行为特征的检测方法，即通过车速、车道横向位置、横摆角、方向盘转角、油门踏板、刹车踏板等车辆行驶信息判断驾驶员是否疲劳。
(4)基于信息融合技术的检测方法，即融合以上两种或多种检测方法。
然而，发明人意识到，上述方法(1)需要在驾驶员身上佩戴专门的仪器才可监测信号，操作较为复杂。上述方法(2)需要进行复杂的图像处理，操作比较复杂，且数据处理量大，效率较低。上述方法(3)和(4)基于驾驶员操作行为特征的检测方法，由于实时采集数据方便，成本低，易普及，对驾驶员驾驶车辆过程中没有任何干扰，等特点，成为疲劳驾驶检测的重要研究方向。但是传统的基于驾驶员操作行为特征的检测方法一般是每一种检测方法都基于一个特征，特征过于单一，容易出现误报；或者是通过SVM(支持向量机)进行数据训练，SVM训练数据时综合考虑了多个特征信息，但需要训练样本数足够多，实车标定参数一旦固定便不能再修改，对于个体差异不能做适应性调节，因此准确性较低。
发明内容
根据本申请公开的各种实施例，提供一种疲劳驾驶状态检测方法、装置、计算机设备和存储介质。
一种疲劳驾驶状态检测方法，包括：
获取车辆的当前行驶时长和当前行驶速度；
当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
查询预先配置的与所述参考参数对应的判断阈值；
判断所述参考参数值是否符合所述判断阈值；及
当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
一种疲劳驾驶状态检测装置，包括：
时长速度采集模块，用于获取车辆的当前行驶时长和当前行驶速度；
参考参数值采集模块，用于当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
判断阈值查询模块，用于查询预先配置的与所述参考参数对应的判断阈值；
第一判断模块，用于判断所述参考参数值是否符合所述判断阈值；及
判定模块，用于当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
一种计算机设备，包括存储器和一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述一个或多个处理器执行以下步骤：
获取车辆的当前行驶时长和当前行驶速度；
当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
查询预先配置的与所述参考参数对应的判断阈值；
判断所述参考参数值是否符合所述判断阈值；及
当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
获取车辆的当前行驶时长和当前行驶速度；
当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
查询预先配置的与所述参考参数对应的判断阈值；
判断所述参考参数值是否符合所述判断阈值；及
当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域 普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为根据一个或多个实施例中疲劳状态检测方法的应用环境图。
图2为根据一个或多个实施例中疲劳状态检测方法的流程示意图。
图3为根据一个或多个实施例中的判断阈值的配置方式的流程图。
图4为根据另一个或多个实施例中疲劳状态检测方法的流程示意图。
图5为根据一个或多个实施例中疲劳状态检测装置的框图。
图6为根据一个或多个实施例中计算机设备的框图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供的疲劳驾驶状态检测方法，可以应用于如图1所示的应用环境中。车辆终端102与车辆上的各个控制器104进行通信，以采集车辆的当前行驶时长、当前行驶速度以及参考参数值，当车辆终端102判定车辆的当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取车辆的参考参数值，然后查询到正常行驶时所配置的与参考参数对应的判断阈值，从而可以判断参考参数值是否符合判断阈值，若是，则可以判定驾驶车辆的驾驶员处于疲劳驾驶状态。上述疲劳驾驶状态检测方法，引入自定义的判断阈值，即每个驾驶员在使用时均设置正常行驶时所配置的与参考参数对应的判断阈值，这样充分考虑了个体差异，保证了判断阈值的准确性，从而可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率。车辆终端102可以但不限于是安装在车辆上的各种智能控制设备，包括个人计算机、笔记本电脑、智能手机、平板电脑。控制器则是车辆上安装的用于采集车辆信息的各种控制器。
在其中一个实施例中，如图2所示，提供了一种疲劳驾驶状态检测方法，以该方法应用于图1中的车辆终端为例进行说明，包括以下步骤：
S202：获取车辆的当前行驶时长和当前行驶速度。
具体地，车辆的当前行驶时长可以是车辆终端所记录的车辆电源为开启状态的时长，车辆的当前行驶速度则是车辆终端通过速度传感器所采集到的。车辆终端可以在车辆行驶后实时采集车辆的当前行驶时长和当前行驶速度。
在其中一个实施例中，获取车辆的当前行驶时长和当前行驶速度包括：获取电源状态以及车辆的当前行驶速度；统计电源状态处于开启状态时的车辆的当前行驶时长。这样在统计行驶时长时，考虑到了电源状态，提高行驶时长统计的准确性，这样充分考虑到车速较低时疲劳驾驶危险指数相对较低，且可能存在疲劳驾驶的误报，进而也提高了疲劳驾驶状态检测的准确性。
在实际使用过程中，车辆终端在统计驾驶时长时可以包括以下三种情况：第一，电源 状态ON档，车速有效且车速大于等于预设速度，例如5km/h时开始统计驾驶时长。第二，电源状态变为ACC或OFF档时驾驶时长清零，再次满足条件重新计时。第三，电源状态为ON档，车速无效或车速小于预设速度，例如5km/h超过车辆停止计时时长，即VehicleStopTimer时驾驶时长清零，再次满足条件重新计时。
S204：当当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取车辆的参考参数值。
具体地，预设要求是指需要进行疲劳驾驶状态判断的最低要求，在该实施例中，考虑车速较低时疲劳驾驶危险指数相对较低且可能会存在疲劳驾驶的误报给用户带来不好的体验，车速可以限定在30km/h以上，在其他实施例中可以限定为其他速度，当前行驶时长则可以限制为半小时等，在其他实施例中可以限定为其他时长。
预先确定的参考参数是通过采集驾驶人疲劳驾驶数据，对采集的数据进行分析，提取与疲劳驾驶相关的参数，该参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数，在其他实施例中，上述参考参数还可以替换为其他的与疲劳驾驶有关的有效的特征参考参数。车辆终端当前行驶时长和当前行驶速度满足预设要求时可以采集车速、方向盘转角、加速踏板、刹车踏板及横摆角速度等基础数据并加以进行统计分析得到参考参数值。例如，车辆终端中可以预先存储有计算参考参数值的公式，这样在车辆终端采集得到上述基础数据后，则根据该公式求得对应的参考参数值。
S206：查询预先配置的与参考参数对应的判断阈值。
具体地，判断阈值是驾驶员在使用车辆之前或者是使用过程中，在车辆正常行驶时所配置的与参考参数对应的判断阈值，例如在驾驶员初次使用车辆时，则可以通过在控制屏上的对应的配置开关来开启配置状态，并进行配置。当驾驶员使用多次后，则可以通过在控制屏上的对应的配置开关来开启配置状态，以更新原来的配置，例如正常驾驶一段时间后，统计N秒数据后，使用该段数据更新判断阈值，并将所配置的判断阈值与对应的驾驶员的头像进行关联存储。
车辆终端可以在当前行驶时长和当前行驶速度满足预设要求时，则查询预先配置的与参考参数对应的判断阈值，且该查询步骤和上述的获取参考参数值的步骤可以是并行的。此外，车辆终端可以首先采集驾驶员的头像，然后将所采集的头像与预先存储的头像进行匹配，获取匹配度最大的头像对应的判断阈值作为本次判断所使用的阈值。
S208：判断参考参数值是否符合判断阈值。
S210：当参考参数值符合基于判断阈值的判断条件时，则驾驶车辆的驾驶员处于疲劳驾驶状态。
具体地，判断参考参数值是否符合判断阈值则可以是车辆终端根据预设的判断逻辑进行，例如，单位时间内特征指标油门刹车总操作时间占比小于第二阈值P th，单位时间内方向盘最大不动时间大于第三阈值T θ_th，单位时间内油门和刹车总次数小于第四阈值Freq th，则确定驾驶车辆的驾驶员处于疲劳驾驶状态，只要有一个不满足，则判定驾驶车辆的驾驶 员未处于疲劳驾驶状态。
上述疲劳驾驶状态检测方法，引入自定义的判断阈值，即每个驾驶员在使用时均设置正常行驶时所配置的与参考参数对应的判断阈值，这样充分考虑了个体差异，保证了判断阈值的准确性，从而可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率，此外只有当前行驶时长和当前行驶速度满足预设要求时才会进行检测，这样充分考虑到车速较低时疲劳驾驶危险指数相对较低，且可能存在疲劳驾驶的误报，进而也提高了疲劳驾驶状态检测的准确性。
在其中一个实施例中，判断阈值的配置方式包括：接收驾驶员输入的阈值配置指令；根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与参考参数对应；根据行驶参数值计算得到判断阈值并配置。
具体的判断阈值的配置包括两个阶段：第一阶段是驾驶员初次驾驶车辆，则强制进行配置，第二阶段是驾驶员使用了一段时间后，则需要更新第一阶段或者是上一次设置的判断阈值。上述两个阶段驾驶员可以通过点击控制屏幕上的判断阈值配置按钮来开启对判断阈值的配置，其中针对第二阶段车辆终端可以进行判断，例如判断判断阈值的使用时长是否大于预设时长，例如大于半年或者是一年，若是，则提示进行判断阈值的更新，从而驾驶员可以通过点击控制屏幕上的判断阈值配置按钮来开启对判断阈值的配置。
具体地，判断阈值的配置方式可以是，车辆终端接收到驾驶员输入的阈值配置指令，然后车辆终端开启配置状态，这样车辆终端采集正常行驶状态下的车速，方向盘转角，加速踏板，刹车踏板及横摆角速度等基础数据，然后进行统计分析得到与参考参数对应的行驶参数值，最后根据行驶参数值计算得到判断阈值并配置，例如取行驶参数值中的最大值作为判断阈值等。
以上述例子中的参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数为例进行说明：
在其中一个实施例中，方向盘最大不动时间对应的判断阈值的获取方式包括：获取预设时间段内每次方向盘不动时的方向盘转角；计算相邻两次方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第一时间差；将最大的第一时间差作为方向盘最大不动时间对应的判断阈值。具体地，用户可以自定义设置方向盘转角不动阈值，默认采用阈值θ。其中实际应用中针对方向盘最大不动时间的判断阈值的设置可以包括：针对不同驾驶人，在正常驾驶一定时间时，统计N秒内控制方向盘不动时，方向盘转角θ i值；计算相邻两次方向盘转角的差值Δθ＝|θ i+1-θ i|。以单位时间窗内Δθ中最大值Δθ max作为方向盘转角不动阈值θ。具体设置可在大屏上做自定义设置阈值开关的形式实现。统计N秒数据获取到阈值后关闭设置功能。该实施例中，在获取到方向盘最大不动时间对应的判断阈值时，首先确定方向盘不动阈值，这样再统计单位时间窗内的方向盘的转角的差值小于方向盘不动阈值的第一时间差，可以保证准确性。
在其中一个实施例中，油门刹车总次数对应的判断阈值的配置方式包括：采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；根据油门踏板操作频率以及刹车踏板操作频率得到油门刹车总次数。
在其中一个实施例中，单位时间内油门刹车总操作时间占比对应的判断阈值的配置方式包括：采集单位时间窗内油门开度非零点的时间作为油门操作时间；采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；根据油门操作时间、刹车踏板操作时间以及单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
上述实施例中，驾驶员可以自定义的进行判断阈值的设置，且在设置的时候充分考虑到车辆的性能，从而使得判断阈值更加符合驾驶员和当前车辆，保证了判断阈值的准确性。
在其中一个实施例中，接收驾驶员输入的阈值配置指令之前，还包括：通过图像采集设备采集驾驶员的头像；根据头像判断是否存储有与驾驶员对应的判断阈值；当未存储有与驾驶员对应的判断阈值时，则输出判断阈值配置提示。其中图像采集设备可以是设置在车辆中的可以拍摄到驾驶员头像的设备，车辆终端可以通过位置识别坐在驾驶座位上的驾驶员的头像，该头像至少要包括驾驶员的人脸，且由于判断阈值是与驾驶员的头像进行关联存储的，因此可以根据该人脸查询到对应的判断阈值，若不存在，则说明该驾驶员是初次驾驶车辆，需要进行判断阈值的配置，因此输出判断阈值配置提示。上述实施例中，预先根据图像采集设备所采集的驾驶员的头像判断车辆中是否存储有与驾驶员对应的判断阈值，这样在没有存储时，可以提示驾驶员进行配置，保证了判断阈值的个性化，从而提高可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率。
在其中一个实施例中，根据头像判断是否存储有与驾驶员对应的判断阈值之后，还包括：当存储有与驾驶员对应的判断阈值时，则获取判断阈值的生成时间；根据当前系统时间与生成时间得到判断阈值的使用时长；判断使用时长是否大于预设时长；当使用时长大于预设时长时，则输出判断阈值配置提示。其中生成时间是指判断阈值存储时间，即判断阈值存储到车辆终端的时间，这样车辆终端根据当前系统时间和生成时间就可以得到判断阈值的使用时长。预设时长时用户预先设置的，其可以为半年或者是一年等，当判断阈值的使用时长大于预设时长时，则说明判断阈值的使用时间较长，这样可能会存在因为时间过长导致的判断阈值存在误差，因此车辆终端提示驾驶员重新进行配置。上述实施例中，预先根据图像采集设备所采集的驾驶员的头像判断车辆中是否存储有与驾驶员对应的判断阈值，这样在存储有判断阈值时，还需要获取到判断阈值的使用时长，这样避免因为时间过长导致的判断阈值存在误差的问题，保证了判断阈值的个性化，从而提高可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率。
具体地，请参见图3所示，图3为一个实施例中的判断阈值的配置方式的流程图，在该实施例中，该判断阈值的配置方式主要包括：车辆终端接收驾驶员输入的阈值配置指令，然后车辆终端通过图像采集设备采集驾驶员的头像，并判断车辆终端中是否存储有与所采 集的头像对应的判断阈值，这样当不存在时，则输出判断阈值配置提示，也就说车辆终端一边输出判断阈值配置提示，一边根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与所述参考参数对应；根据行驶参数值计算得到判断阈值并配置。若存在与头像对应的判断阈值时，则车辆终端获取判断阈值的生成时间；根据当前系统时间与生成时间得到判断阈值的使用时长；判断使用时长是否大于预设时长；当使用时长大于预设时长时，则输出判断阈值配置提示，即根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与所述参考参数对应；根据行驶参数值计算得到判断阈值并配置。如果存在判断阈值且判断阈值的使用时长小于等于预设时长，则可以直接输出已存在判断阈值的提示，由用户选择是否再次进行配置。
上述实施例中，预先根据图像采集设备所采集的驾驶员的头像判断车辆中是否存储有与驾驶员对应的判断阈值，这样在存储有判断阈值时，还需要获取到判断阈值的使用时长，这样避免因为时间过长导致的判断阈值存在误差的问题，保证了判断阈值的个性化，从而提高可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率。
在其中一个实施例中，参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。上述实施例中，综合考虑了多个驾驶员行为特征，可以多方面地对疲劳驾驶状态进行判断，提高疲劳驾驶状态的检测准确性。
在其中一个实施例中，方向盘最大不动时间的采集方式包括：统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第二时间差；将最大的第二时间差作为方向盘最大不动时间。
具体地，车辆终端在检测到相邻时间内方向盘转角之差小于阈值θ时，则认为方向盘不动。这样车辆终端统计单位时间窗T内方向盘连续不动时间，取最大值作为方向盘最大不动时间T θ。
上述实施例中，通过统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第二时间差，来实时获取到方向盘最大不动时间，提高准确性。
在其中一个实施例中，油门刹车总次数的采集方式包括：采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；根据油门踏板操作频率以及刹车踏板操作频率得到油门刹车总次数。
具体地，油门开度Trt由非零变为零，认为松开油门一次，单位时间窗T内油门踩踏频率Freq Trt：
Freq Trt＝Freq Trt+1              (1)
刹车踏板Brk由非零变为零，认为松开刹车一次，单位时间窗内刹车踩踏频率Freq Brk
Freq Brk＝Freq Brk+1              (2)
单位时间内油门刹车总操作次数Freq为单位时间内油门操作频率和刹车操作频率之和：
Freq＝Freq Trt+Freq Brk          (3)
上述实施例中，通过油门开度发生变化的频率来表示油门踏板操作频率，通过刹车踏板发生变化的频率来表示刹车踏板操作频率，检测方便，且准确性高。
在其中一个实施例中，单位时间内油门刹车总操作时间占比的采集方式包括：采集单位时间窗内油门开度非零点的时间作为油门操作时间；采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；根据油门操作时间、刹车踏板操作时间以及单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
油门操作时间Trt time通过油门开度非0点统计：
Figure PCTCN2020085627-appb-000001
刹车踏板操作时间Brk time通过制动缸压力的非零点个数表示：
Figure PCTCN2020085627-appb-000002
油门刹车操作总时间Total time：
Total time＝Trt time+Brk time          (6)
油门刹车操作总时间占比P：
Figure PCTCN2020085627-appb-000003
上述实施例中，通过油门开度非零点的时间来表示油门操作时间，通过制动缸压力的非零点的时间来表示刹车踏板操作时间，检测方便，且准确性高。
在其中一个实施例中，判断参考参数值是否符合判断阈值包括：判断方向盘最大不动时间是否大于第一判断阈值，单位时间内油门刹车总操作时间占比是否小于第二判断阈值，油门刹车总次数是否小于第三判断阈值；当方向盘最大不动时间大于第一判断阈值，单位时间内油门刹车总操作时间占比小于第二判断阈值，油门刹车总次数小于第三判断阈值时，则参考参数值符合判断阈值。上述实施例中，综合考虑了多个驾驶员行为特征，可以多方面地对疲劳驾驶状态进行判断，提高疲劳驾驶状态的检测准确性。
具体地，请参见图4，图4为另一个实施例中的疲劳驾驶状态检测方法的流程图，在该实施例中，车辆终端首先采集驾驶时间和车速，若驾驶时间和车速满足预设条件，例如驾驶时间超过T 0，车速大于30km/h情况下，则需要确认驾驶员是否处于疲劳驾驶状态，这样车辆终端采集基础数据，例如方向盘转角、油门和刹车数据等，然后车辆终端计算各个参考参数值，即通过综合方向盘不动时间，油门刹车踏板踩踏频率以及油门刹车总操作时间占比三个特征进行疲劳驾驶的检测。车辆终端可以设置时间窗口T，两个连续时间窗重复率为Tr，也就是说每统计T时间特征得出一次疲劳驾驶状态，前一个T时间和后一个T时间重复时间为T r；相当于每隔时间(T-T r)输出一次疲劳驾驶状态统计单位时间T内的方向盘最大不动时间T θ，油门刹车操作总时间占比P，油门刹车总操作次数Freq，然后车辆终端依次进行如下判断：油门和刹车操作总时间P<P th；油门刹车总操作次数Freq< Freq th；方向盘最大不动时间T θ>T θ_th；若满足以上所有条件认为驾驶员处于疲劳状态。否则，驾驶员未处于疲劳驾驶状态。
上述疲劳驾驶状态检测方法，引入自定义的判断阈值，即每个驾驶员在使用时均设置正常行驶时所配置的与参考参数对应的判断阈值，这样充分考虑了个体差异，保证了判断阈值的准确性，从而可以提高疲劳驾驶状态检测的准确性，进而可以减少误报率，此外只有当前行驶时长和当前行驶速度满足预设要求时才会进行检测，这样充分考虑到车速较低时疲劳驾驶危险指数相对较低，且可能存在疲劳驾驶的误报，进而也提高了疲劳驾驶状态检测的准确性。
应该理解的是，虽然图2-4的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2-4中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在其中一个实施例中，如图5所示，提供了一种疲劳驾驶状态检测装置，包括：时长速度采集模块100、参考参数值采集模块200、判断阈值查询模块300、第一判断模块400和判定模块500，其中：
时长速度采集模块100，用于获取车辆的当前行驶时长和当前行驶速度；
参考参数值采集模块200，用于当当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取车辆的参考参数值；
判断阈值查询模块300，用于查询预先配置的与参考参数对应的判断阈值；
第一判断模块400，用于判断参考参数值是否符合判断阈值；
判定模块500，用于当参考参数值符合基于判断阈值的判断条件时，则驾驶车辆的驾驶员处于疲劳驾驶状态。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
接收模块，用于接收驾驶员输入的阈值配置指令；
参数采集模块，用于根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与参考参数对应；
判断阈值配置模块，用于根据行驶参数值计算得到判断阈值并配置。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
图像采集模块，用于通过图像采集设备采集驾驶员的头像；
第二判断模块，用于根据头像判断是否存储有与驾驶员对应的判断阈值；
第一提示模块，用于当未存储有与驾驶员对应的判断阈值时，则输出判断阈值配置提示。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
生成时间获取模块，用于当存储有与驾驶员对应的判断阈值时，则获取判断阈值的生成时间；
使用时长获取模块，用于根据当前系统时间与生成时间得到判断阈值的使用时长；
第三判断模块，用于判断使用时长是否大于预设时长；
第二提示模块，用于当使用时长大于预设时长时，则输出判断阈值配置提示。
在其中一个实施例中，参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
方向盘转角获取模块，用于获取预设时间段内每次方向盘不动时的方向盘转角；
方向盘不动阈值获取模块，用于计算相邻两次方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；
第一时间差统计模块，用于统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第一时间差；
第一配置模块，用于将最大的第一时间差作为方向盘最大不动时间对应的判断阈值。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
第二时间差统计模块，用于统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第二时间差；
第二配置模块，用于将最大的第二时间差作为方向盘最大不动时间。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
油门踏板操作频率采集模块，用于采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；
刹车踏板操作频率采集模块，用于采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；
油门刹车总次数计算模块，用于根据油门踏板操作频率以及刹车踏板操作频率得到油门刹车总次数。
在其中一个实施例中，上述的疲劳驾驶状态检测装置还可以包括：
油门操作时间采集模块，用于采集单位时间窗内油门开度非零点的时间作为油门操作时间；
刹车踏板操作时间采集模块，用于采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；
占比统计模块，用于根据油门操作时间、刹车踏板操作时间以及单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
在其中一个实施例中，上述的第一判断模块400可以包括：
判断单元，用于判断方向盘最大不动时间是否大于第一判断阈值，单位时间内油门刹 车总操作时间占比是否小于第二判断阈值，油门刹车总次数是否小于第三判断阈值；
输出单元，用于当方向盘最大不动时间大于第一判断阈值，单位时间内油门刹车总操作时间占比小于第二判断阈值，油门刹车总次数小于第三判断阈值时，则参考参数值符合判断阈值。
在其中一个实施例中，上述的时长速度采集模块100可以包括：
第一获取单元，用于获取电源状态以及车辆的当前行驶速度；
第二获取单元，用于统计电源状态处于开启状态时的车辆的当前行驶时长。
关于疲劳驾驶状态检测装置的具体限定可以参见上文中对于疲劳驾驶状态检测方法的限定，在此不再赘述。上述疲劳驾驶状态检测装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在一个实施例中，提供了一种计算机设备，该计算机设备可以是终端，其内部结构图可以如图6所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口、显示屏和输入装置。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机可读指令。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种疲劳驾驶状态检测方法。该计算机设备的显示屏可以是液晶显示屏或者电子墨水显示屏，该计算机设备的输入装置可以是显示屏上覆盖的触摸层，也可以是计算机设备外壳上设置的按键、轨迹球或触控板，还可以是外接的键盘、触控板或鼠标等。
本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
一种计算机设备，包括存储器和一个或多个处理器，存储器中储存有计算机可读指令，计算机可读指令被处理器执行时，使得一个或多个处理器执行以下步骤：获取车辆的当前行驶时长和当前行驶速度；当当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取车辆的参考参数值；查询预先配置的与参考参数对应的判断阈值；判断参考参数值是否符合判断阈值；当参考参数值符合基于判断阈值的判断条件时，则驾驶车辆的驾驶员处于疲劳驾驶状态。
在一个实施例中，处理器执行计算机可读指令时所涉及的判断阈值的配置方式包括：接收驾驶员输入的阈值配置指令；根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与参考参数对应；根据行驶参数值计算得到判断阈值并配置。
在一个实施例中，处理器执行计算机可读指令时所实现的接收驾驶员输入的阈值配置 指令之前，还包括：通过图像采集设备采集驾驶员的头像；根据头像判断是否存储有与驾驶员对应的判断阈值；当未存储有与驾驶员对应的判断阈值时，则输出判断阈值配置提示。
在一个实施例中，处理器执行计算机可读指令时所实现的根据头像判断是否存储有与驾驶员对应的判断阈值之后，还包括：当存储有与驾驶员对应的判断阈值时，则获取判断阈值的生成时间；根据当前系统时间与生成时间得到判断阈值的使用时长；判断使用时长是否大于预设时长；当使用时长大于预设时长时，则输出判断阈值配置提示。
在一个实施例中，处理器执行计算机可读指令时所涉及的参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。
在一个实施例中，处理器执行计算机可读指令时所涉及的方向盘最大不动时间对应的判断阈值的获取方式包括：获取预设时间段内每次方向盘不动时的方向盘转角；计算相邻两次方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第一时间差；将最大的第一时间差作为方向盘最大不动时间对应的判断阈值。
在一个实施例中，处理器执行计算机可读指令时所涉及的方向盘最大不动时间的采集方式包括：统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第二时间差；将最大的第二时间差作为方向盘最大不动时间。
在一个实施例中，处理器执行计算机可读指令时所涉及的油门刹车总次数的采集方式包括：采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；根据油门踏板操作频率以及刹车踏板操作频率得到油门刹车总次数。
在一个实施例中，处理器执行计算机可读指令时所涉及的单位时间内油门刹车总操作时间占比的采集方式包括：采集单位时间窗内油门开度非零点的时间作为油门操作时间；采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；根据油门操作时间、刹车踏板操作时间以及单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
在一个实施例中，处理器执行计算机可读指令时所实现的判断参考参数值是否符合判断阈值包括：判断方向盘最大不动时间是否大于第一判断阈值，单位时间内油门刹车总操作时间占比是否小于第二判断阈值，油门刹车总次数是否小于第三判断阈值；当方向盘最大不动时间大于第一判断阈值，单位时间内油门刹车总操作时间占比小于第二判断阈值，油门刹车总次数小于第三判断阈值时，则参考参数值符合判断阈值。
在一个实施例中，处理器执行计算机可读指令时所实现的获取车辆的当前行驶时长和当前行驶速度包括：获取电源状态以及车辆的当前行驶速度；统计电源状态处于开启状态时的车辆的当前行驶时长。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：获取车辆的当前行驶 时长和当前行驶速度；当当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取车辆的参考参数值；查询预先配置的与参考参数对应的判断阈值；判断参考参数值是否符合判断阈值；当参考参数值符合基于判断阈值的判断条件时，则驾驶车辆的驾驶员处于疲劳驾驶状态。
在一个实施例中，计算机可读指令被处理器执行时所涉及的判断阈值的配置方式包括：接收驾驶员输入的阈值配置指令；根据阈值配置指令采集车辆行驶时的行驶参数值，行驶参数值与参考参数对应；根据行驶参数值计算得到判断阈值并配置。
在一个实施例中，计算机可读指令被处理器执行时所实现的接收驾驶员输入的阈值配置指令之前，还包括：通过图像采集设备采集驾驶员的头像；根据头像判断是否存储有与驾驶员对应的判断阈值；当未存储有与驾驶员对应的判断阈值时，则输出判断阈值配置提示。
在一个实施例中，计算机可读指令被处理器执行时所实现的根据头像判断是否存储有与驾驶员对应的判断阈值之后，还包括：当存储有与驾驶员对应的判断阈值时，则获取判断阈值的生成时间；根据当前系统时间与生成时间得到判断阈值的使用时长；判断使用时长是否大于预设时长；当使用时长大于预设时长时，则输出判断阈值配置提示。
在一个实施例中，计算机可读指令被处理器执行时所涉及的参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。
在一个实施例中，计算机可读指令被处理器执行时所涉及的方向盘最大不动时间对应的判断阈值的获取方式包括：获取预设时间段内每次方向盘不动时的方向盘转角；计算相邻两次方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第一时间差；将最大的第一时间差作为方向盘最大不动时间对应的判断阈值。
在一个实施例中，计算机可读指令被处理器执行时所涉及的方向盘最大不动时间的采集方式包括：统计单位时间窗内方向盘的转角的差值小于方向盘不动阈值的第二时间差；将最大的第二时间差作为方向盘最大不动时间。
在一个实施例中，计算机可读指令被处理器执行时所涉及的油门刹车总次数的采集方式包括：采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；根据油门踏板操作频率以及刹车踏板操作频率得到油门刹车总次数。
在一个实施例中，计算机可读指令被处理器执行时所涉及的单位时间内油门刹车总操作时间占比的采集方式包括：采集单位时间窗内油门开度非零点的时间作为油门操作时间；采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；根据油门操作时间、刹车踏板操作时间以及单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
在一个实施例中，计算机可读指令被处理器执行时所实现的判断参考参数值是否符合 判断阈值包括：判断方向盘最大不动时间是否大于第一判断阈值，单位时间内油门刹车总操作时间占比是否小于第二判断阈值，油门刹车总次数是否小于第三判断阈值；当方向盘最大不动时间大于第一判断阈值，单位时间内油门刹车总操作时间占比小于第二判断阈值，油门刹车总次数小于第三判断阈值时，则参考参数值符合判断阈值。
在一个实施例中，计算机可读指令被处理器执行时所实现的获取车辆的当前行驶时长和当前行驶速度包括：获取电源状态以及车辆的当前行驶速度；统计电源状态处于开启状态时的车辆的当前行驶时长。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种疲劳驾驶状态检测方法，包括：
    获取车辆的当前行驶时长和当前行驶速度；
    当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
    查询预先配置的与所述参考参数对应的判断阈值；
    判断所述参考参数值是否符合所述判断阈值；及
    当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
  2. 根据权利要求1所述的方法，其特征在于，所述判断阈值的配置方式包括：
    接收驾驶员输入的阈值配置指令；
    根据所述阈值配置指令采集所述车辆行驶时的行驶参数值，所述行驶参数值与所述参考参数对应；及
    根据所述行驶参数值计算得到判断阈值并配置。
  3. 根据权利要求2所述的方法，其特征在于，所述接收驾驶员输入的阈值配置指令之前，还包括：
    通过图像采集设备采集驾驶员的头像；
    根据所述头像判断是否存储有与所述驾驶员对应的判断阈值；及
    当未存储有与所述驾驶员对应的判断阈值时，则输出判断阈值配置提示；
    当存储有与所述驾驶员对应的判断阈值时，则获取所述判断阈值的生成时间；
    根据当前系统时间与所述生成时间得到所述判断阈值的使用时长；
    判断所述使用时长是否大于预设时长；及
    当所述使用时长大于预设时长时，则输出判断阈值配置提示。
  4. 根据权利要求1至3任意一项所述的方法，其特征在于，所述参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。
  5. 根据权利要求4所述的方法，其特征在于，所述方向盘最大不动时间对应的判断阈值的获取方式包括：
    获取预设时间段内每次所述方向盘不动时的方向盘转角；
    计算相邻两次所述方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；
    统计单位时间窗内所述方向盘的转角的差值小于所述方向盘不动阈值的第一时间差；及
    将最大的所述第一时间差作为所述方向盘最大不动时间对应的判断阈值。
  6. 根据权利要求5所述的方法，其特征在于，所述方向盘最大不动时间的采集方式包括：
    统计单位时间窗内所述方向盘的转角的差值小于所述方向盘不动阈值的第二时间差；及
    将最大的所述第二时间差作为所述方向盘最大不动时间。
  7. 根据权利要求4所述的方法，其特征在于，所述油门刹车总次数的采集方式包括：
    采集单位时间窗内油门开度发生变化的频率，作为油门踏板操作频率；
    采集单位时间窗内刹车踏板发生变化的频率，作为刹车踏板操作频率；及
    根据所述油门踏板操作频率以及所述刹车踏板操作频率得到油门刹车总次数。
  8. 根据权利要求4所述的方法，其特征在于，所述单位时间内油门刹车总操作时间占比的采集方式包括：
    采集单位时间窗内油门开度非零点的时间作为油门操作时间；
    采集单位时间窗内制动缸压力的非零点的时间作为刹车踏板操作时间；及
    根据所述油门操作时间、所述刹车踏板操作时间以及所述单位时间窗的时间长度计算得到单位时间内油门刹车总操作时间占比。
  9. 根据权利要求4所述的方法，其特征在于，所述判断所述参考参数值是否符合所述判断阈值包括：
    判断所述方向盘最大不动时间是否大于第一判断阈值，所述单位时间内油门刹车总操作时间占比是否小于第二判断阈值，所述油门刹车总次数是否小于第三判断阈值；及
    当所述方向盘最大不动时间大于第一判断阈值，所述单位时间内油门刹车总操作时间占比小于第二判断阈值，所述油门刹车总次数小于第三判断阈值时，则所述参考参数值符合判断阈值。
  10. 根据权利要求1至3任意一项所述的方法，其特征在于，所述获取车辆的当前行驶时长和当前行驶速度包括：
    获取电源状态以及车辆的当前行驶速度；及
    统计所述电源状态处于开启状态时的车辆的当前行驶时长。
  11. 一种疲劳驾驶状态检测装置，包括：
    时长速度采集模块，用于获取车辆的当前行驶时长和当前行驶速度；
    参考参数值采集模块，用于当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
    判断阈值查询模块，用于查询预先配置的与所述参考参数对应的判断阈值；
    第一判断模块，用于判断所述参考参数值是否符合所述判断阈值；及
    判定模块，用于当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
  12. 根据权利要求11所述的装置，其特征在于，所述装置还包括：
    接收模块，用于接收驾驶员输入的阈值配置指令；
    参数采集模块，用于根据所述阈值配置指令采集所述车辆行驶时的行驶参数值，所述 行驶参数值与所述参考参数对应；及
    判断阈值配置模块，用于根据所述行驶参数值计算得到判断阈值并配置。
  13. 一种计算机设备，包括存储器及一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取车辆的当前行驶时长和当前行驶速度；
    当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
    查询预先配置的与所述参考参数对应的判断阈值；
    判断所述参考参数值是否符合所述判断阈值；及
    当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
  14. 根据权利要求13所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时所涉及的所述判断阈值的配置方式包括：
    接收驾驶员输入的阈值配置指令；
    根据所述阈值配置指令采集所述车辆行驶时的行驶参数值，所述行驶参数值与所述参考参数对应；及
    根据所述行驶参数值计算得到判断阈值并配置。
  15. 根据权利要求14所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时所实现的所述接收驾驶员输入的阈值配置指令之前，还包括：
    通过图像采集设备采集驾驶员的头像；
    根据所述头像判断是否存储有与所述驾驶员对应的判断阈值；及
    当未存储有与所述驾驶员对应的判断阈值时，则输出判断阈值配置提示；
    当存储有与所述驾驶员对应的判断阈值时，则获取所述判断阈值的生成时间；
    根据当前系统时间与所述生成时间得到所述判断阈值的使用时长；
    判断所述使用时长是否大于预设时长；及
    当所述使用时长大于预设时长时，则输出判断阈值配置提示。
  16. 根据权利要求13至15任意一项所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时所实现的所述参考参数包括但不限于单位时间内油门刹车总操作时间占比、方向盘最大不动时间以及油门刹车总次数。
  17. 根据权利要求16所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时所涉及的所述方向盘最大不动时间对应的判断阈值的获取方式包括：
    获取预设时间段内每次所述方向盘不动时的方向盘转角；
    计算相邻两次所述方向盘不动时的方向盘转角的差值，并获取到最大差值作为方向盘不动阈值；
    统计单位时间窗内所述方向盘的转角的差值小于所述方向盘不动阈值的第一时间差；及
    将最大的所述第一时间差作为所述方向盘最大不动时间对应的判断阈值。
  18. 根据权利要求17所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时所涉及的所述方向盘最大不动时间的采集方式包括：
    统计单位时间窗内所述方向盘的转角的差值小于所述方向盘不动阈值的第二时间差；及
    将最大的所述第二时间差作为所述方向盘最大不动时间。
  19. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取车辆的当前行驶时长和当前行驶速度；
    当所述当前行驶时长和当前行驶速度满足预设要求时，则根据预先确定的参考参数，获取所述车辆的参考参数值；
    查询预先配置的与所述参考参数对应的判断阈值；
    判断所述参考参数值是否符合所述判断阈值；及
    当所述参考参数值符合基于所述判断阈值的判断条件时，则驾驶所述车辆的驾驶员处于疲劳驾驶状态。
  20. 根据权利要求19所述的存储介质，其特征在于，所述计算机可读指令被一个或多个处理器执行时所涉及的所述判断阈值的配置方式包括：
    接收驾驶员输入的阈值配置指令；
    根据所述阈值配置指令采集所述车辆行驶时的行驶参数值，所述行驶参数值与所述参考参数对应；及
    根据所述行驶参数值计算得到判断阈值并配置。
PCT/CN2020/085627 2020-04-20 2020-04-20 疲劳驾驶状态检测方法、装置、计算机设备和存储介质 WO2021212274A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/085627 WO2021212274A1 (zh) 2020-04-20 2020-04-20 疲劳驾驶状态检测方法、装置、计算机设备和存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/085627 WO2021212274A1 (zh) 2020-04-20 2020-04-20 疲劳驾驶状态检测方法、装置、计算机设备和存储介质

Publications (1)

Publication Number Publication Date
WO2021212274A1 true WO2021212274A1 (zh) 2021-10-28

Family

ID=78271022

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/085627 WO2021212274A1 (zh) 2020-04-20 2020-04-20 疲劳驾驶状态检测方法、装置、计算机设备和存储介质

Country Status (1)

Country Link
WO (1) WO2021212274A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112528919A (zh) * 2020-12-21 2021-03-19 苏州中科华影健康科技有限公司 一种疲劳驾驶检测方法及装置、计算机可读介质
CN114084084A (zh) * 2021-11-17 2022-02-25 一汽解放汽车有限公司 挂车连接状态判断方法、装置计算机设备和存储介质
CN114148339A (zh) * 2022-01-17 2022-03-08 潍柴动力股份有限公司 一种不良驾驶预警方法和装置
CN114435373A (zh) * 2022-03-16 2022-05-06 一汽解放汽车有限公司 疲劳驾驶检测方法、装置、计算机设备和存储介质
CN114526930A (zh) * 2022-03-09 2022-05-24 河南职业技术学院 一种智能网联汽车故障检测方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105261153A (zh) * 2015-11-03 2016-01-20 北京奇虎科技有限公司 车辆行驶监控方法和装置
CN106114515A (zh) * 2016-06-29 2016-11-16 北京奇虎科技有限公司 汽车驾驶行为提醒方法及系统
CN108577869A (zh) * 2018-04-29 2018-09-28 武汉理工大学 基于驾驶指纹的驾驶疲劳监测方法与系统
WO2020054534A1 (ja) * 2018-09-10 2020-03-19 マツダ株式会社 車両の制御方法、車両システム及び車両の制御装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105261153A (zh) * 2015-11-03 2016-01-20 北京奇虎科技有限公司 车辆行驶监控方法和装置
CN106114515A (zh) * 2016-06-29 2016-11-16 北京奇虎科技有限公司 汽车驾驶行为提醒方法及系统
CN108577869A (zh) * 2018-04-29 2018-09-28 武汉理工大学 基于驾驶指纹的驾驶疲劳监测方法与系统
WO2020054534A1 (ja) * 2018-09-10 2020-03-19 マツダ株式会社 車両の制御方法、車両システム及び車両の制御装置

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112528919A (zh) * 2020-12-21 2021-03-19 苏州中科华影健康科技有限公司 一种疲劳驾驶检测方法及装置、计算机可读介质
CN114084084A (zh) * 2021-11-17 2022-02-25 一汽解放汽车有限公司 挂车连接状态判断方法、装置计算机设备和存储介质
CN114084084B (zh) * 2021-11-17 2023-12-26 一汽解放汽车有限公司 挂车连接状态判断方法、装置计算机设备和存储介质
CN114148339A (zh) * 2022-01-17 2022-03-08 潍柴动力股份有限公司 一种不良驾驶预警方法和装置
CN114148339B (zh) * 2022-01-17 2024-05-17 潍柴动力股份有限公司 一种不良驾驶预警方法和装置
CN114526930A (zh) * 2022-03-09 2022-05-24 河南职业技术学院 一种智能网联汽车故障检测方法及系统
CN114526930B (zh) * 2022-03-09 2024-03-26 河南职业技术学院 一种智能网联汽车故障检测方法及系统
CN114435373A (zh) * 2022-03-16 2022-05-06 一汽解放汽车有限公司 疲劳驾驶检测方法、装置、计算机设备和存储介质
CN114435373B (zh) * 2022-03-16 2023-12-22 一汽解放汽车有限公司 疲劳驾驶检测方法、装置、计算机设备和存储介质

Similar Documents

Publication Publication Date Title
WO2021212274A1 (zh) 疲劳驾驶状态检测方法、装置、计算机设备和存储介质
CN111741884B (zh) 交通遇险和路怒症检测方法
EP3033999B1 (en) Apparatus and method for determining the state of a driver
CN110765807B (zh) 驾驶行为分析、处理方法、装置、设备和存储介质
US8035496B2 (en) Method and computer program for identification of inattentiveness by the driver of a vehicle
CN109716411A (zh) 用以监测驾驶员的活动水平的方法和设备
WO2021017329A1 (zh) 一种检测驾驶员分心的方法及装置
EP2380766B1 (en) Impaired Operation Detection Method and Impaired Operation Detection System
WO2022078077A1 (zh) 驾驶风险的预警方法、装置、计算设备及存储介质
CN106218405A (zh) 疲劳驾驶监控方法及云端服务器
CN111439270A (zh) 疲劳驾驶状态检测方法、装置、计算机设备和存储介质
WO2008114839A1 (ja) 覚醒度判定装置及び覚醒度判定方法
JP2020531929A (ja) テレメトリデータおよびウェアラブルセンサデータを用いたレーシングデータ分析のためのシステムおよび方法
CN111369786B (zh) 乘车出行的监控方法、车辆、服务器、系统和设备
WO2008133746A1 (en) Systems and methods for detecting unsafe conditions
CN108688680A (zh) 通过驾驶员简档增强车辆对x应用
CN113022578B (zh) 基于车辆运动信息乘客提醒方法、系统、车辆及存储介质
WO2021159359A1 (zh) 无人驾驶车辆操作系统的数据采集方法和装置
US20220292613A1 (en) System and method for assessing device usage
US20220089086A1 (en) Techniques for automatically reducing annoyance levels of drivers when using driver monitoring systems
Ahadi-Sarkani et al. ADAS-RL: Adaptive vector scaling reinforcement learning for human-in-the-loop lane departure warning
CN117087678A (zh) 一种驾驶行为分析方法、装置、电子设备和存储介质
WO2023230103A1 (en) Deep learning anomaly detection systems for responding to emergency events in vehicles
CN116563828A (zh) 驾驶员状态实时监控方法、装置、电子设备及存储介质
CN114037978A (zh) 一种疲劳度检测方法、装置、电子设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20932578

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20932578

Country of ref document: EP

Kind code of ref document: A1