阅读说明：本技术 一种基于自动驾驶机器人的自动驾驶汽车舒适度测试评价方法和系统 (Automatic driving automobile comfort level test evaluation method and system based on automatic driving robot ) 是由 杨智博 陈晓东 孙海鹏 吴向亮 王磊 于 2021-07-20 设计创作，主要内容包括：本发明公开了一种基于自动驾驶机器人的自动驾驶汽车舒适度测试评价方法和系统,包括：通过自动驾驶机器人控制车辆进行预定的运动,同时记录受试驾驶员在整个预定车辆运动过程的心电数据和自动驾驶机器人控制的车辆状态客观信息；驾驶员根据主观评价标准进行主观舒适度感受评价；对心电信号进行R波提取,时域计算,得到表征心率变异性的时域指标SDNN和RMSSD,方差分析,得到显著差异性、能够表征舒适性的客观数据；根据客观数据和主观评价数据,进行交叉比对,得到符合驾驶员舒适度的最佳车辆运动状态参数。本发明对于车辆运动状态舒适度的控制具有很高的精度,具备很强的可重复性和易用性,并且对不同工况有很好的兼容性。(The invention discloses a method and a system for testing and evaluating the comfort level of an automatic driving automobile based on an automatic driving robot, wherein the method comprises the following steps: the automatic driving robot controls the vehicle to perform preset movement, and records the electrocardiogram data of the tested driver in the whole preset vehicle movement process and the vehicle state objective information controlled by the automatic driving robot; the driver carries out subjective comfort feeling evaluation according to subjective evaluation criteria; performing R wave extraction and time domain calculation on the electrocardiosignals to obtain time domain indexes SDNN and RMSSD representing heart rate variability, and performing variance analysis to obtain objective data with remarkable difference and capability of representing comfort; and performing cross comparison according to the objective data and the subjective evaluation data to obtain the optimal vehicle motion state parameter which accords with the comfort level of the driver. The method has high precision for controlling the comfort level of the vehicle motion state, strong repeatability and usability, and good compatibility to different working conditions.)

1. An automatic driving automobile comfort level test evaluation method based on an automatic driving robot comprises the following steps:

selecting a plurality of tested drivers, and preparing to adopt an electrocardiosignal physiological test recorder to collect electrocardiosignals;

acquiring electrocardiosignals of a tested driver, and controlling the vehicle to perform preset movement through the automatic driving robot; recording electrocardio data of a tested driver in the whole preset vehicle motion process and vehicle state information controlled by the automatic driving robot;

after finishing the preset vehicle movement, the driver carries out subjective comfort feeling evaluation according to the subjective evaluation table;

according to the electrocardiosignals obtained in the second step, R wave extraction is carried out on the electrocardiosignals of the tested driver, and time domain calculation is carried out on the processing result to obtain time domain indexes SDNN and RMSSD representing heart rate variability;

carrying out variance analysis according to the physiological data SDNN and the RMSSD obtained in the step four to obtain objective data which have obvious difference and can represent comfort;

and step six, carrying out cross comparison according to the objective data obtained in the step five and the subjective evaluation data obtained in the step three to obtain the optimal vehicle motion state parameter which accords with the comfort level of the driver.

2. The autopilot-based autopilot comfort test evaluation method of claim 1 comprising:

the subjective scoring benchmark of the subjective evaluation table in the third step is shown in the scoring benchmark in the following table 1, and after each experiment is completed, a driver performs detailed scoring according to subjective feeling to give subjective comfort evaluation:

TABLE 1 Scoring standards

3. The autopilot-based autopilot comfort test evaluation method of claim 1 comprising:

analyzing the objective data obtained in the third step through programming calculation, and analyzing heart rate variability time domain indexes SDNN and RMSSD extracted from the electrocardio data; and analyzing the change conditions of the SDNN and the RMSSD in combination with the subjective evaluation to obtain corresponding comfortable motion state parameters.

4. The autopilot-based autopilot comfort test evaluation method of claim 3 comprising: the corresponding comfort motion state parameters refer to the maximum longitudinal deceleration and the maximum transverse acceleration parameters in the comfort state.

5. The autopilot-based autopilot comfort test evaluation method of claim 1 comprising:

step two, the vehicle state information comprises target acceleration and deceleration, speed and lane change time; the full-automatic driving operation state of the vehicle is simulated through the automatic driving robot in the whole vehicle moving process.

6. The autopilot-based autopilot comfort test evaluation method of claim 1 comprising: the repeated acceleration test precision of the automatic driving automobile is controlled within 0.5m/s 2.

7. The utility model provides an autopilot car comfort level test system based on autopilot, characterized by includes:

a vehicle having a cockpit;

the automatic driving robot is rigidly and fixedly connected in the vehicle and used for sensing and controlling the vehicle to move according to a set motion state;

the physiological data acquisition equipment is fixed on a driver through an electrocardio sensor and acquires electrocardiosignals through the electrocardio sensor.

8. The autopilot-based autopilot comfort testing system of claim 7 wherein the autopilot further comprises:

the pose and motion sensing system of the automatic driving robot is used for sensing the motion state and position of the vehicle and calculating control instruction information by comparing the actual motion state with the target motion state set in advance;

the automatic driving robot control system is used for sending the control instruction information to the automatic driving robot execution system;

and the automatic driving robot executing system is used for controlling the accelerator, the brake and the steering of the vehicle so as to enable the vehicle to move according to the set motion state.

9. The autopilot-based autopilot comfort testing system of claim 7 or 8 wherein the physiological data collection device further comprises a data collection analysis system; the electrocardiosignals are input into the data acquisition and analysis system and are output for analysis after time synchronization of the electrocardiosignals and the vehicle motion state information output by the automatic driving robot.

10. The autopilot-based autopilot comfort testing system of claim 1 wherein the physiological data collection device is a physiological test recorder and the test electrodes are arranged using a ii-lead method.

Technical Field

The invention relates to the technical field of testing of passenger comfort degree of an automatic driving automobile at L2 level and above, in particular to a testing and evaluating method for passenger comfort degree when the vehicle motion state is changed in an automatic driving state, and particularly relates to a method and a system for testing and evaluating the comfort degree of the automatic driving automobile based on an automatic driving robot.

Background

Currently, the automatic driving technology has become one of the leading technologies of automobiles, and the SAE (society of automotive engineering) in the united states divides the automatic driving technology into 6 grades: the 0-level system only has an alarm reminding function or an instant control function for the vehicle, namely the system only gives an alarm or instantly controls the vehicle to the driver under the dangerous condition, such as FCW, AEB, LKA and other functions, and the main bodies of vehicle control, environment monitoring and system response are the driver; the level 1 system only performs continuous control on a single direction of the vehicle, such as transverse direction or longitudinal direction, such as ACC, LCA, APA and the like. The vehicle control main body is a person and a driver, and the environment monitoring and system responding main body is the driver; the 2-level system can simultaneously and continuously control the vehicle in the transverse direction and the longitudinal direction, the vehicle control main body is the system, and the environment monitoring and the system response are both drivers; the 3-level system is conditional automatic driving, and the main bodies of vehicle control and environment monitoring are all systems; the 4-level system is highly automatic driving, and the main bodies of vehicle control, environment monitoring and system response are all systems, but the mode of controlling the vehicle by the driver is also available; the level 5 system is fully autonomous, i.e., unmanned, wherein the level L2 and above autonomous systems control the motion state of the vehicle as a whole rather than the human driver, which brings about a problem as to whether the occupant in the vehicle is comfortable when the vehicle controlled by the autonomous system performs a change of the motion state, such as braking, lane changing, etc.

The passenger travelling comfort that indicates in this patent application is when autodrive vehicle motion state changes, human physiology, psychological factor and the comprehensive experience of state interaction of traveling. When the motion state of the vehicle changes, different degrees of abrupt changes of acceleration and speed are generated, so that various sensory stimuli to passengers are generated, and the comfort degree is changed. At present, most of evaluation on the comfort of a driver is a subjective evaluation mode, and the evaluation method has the defects of high precision, large individual difference and poor experiment repeatability. At present, a comfort evaluation method based on the combination of objective physiological signals and subjective evaluation does not exist, so that the comfort of a driver cannot be expressed in a form of objective quantitative indexes. Therefore, how to objectively and accurately judge the comfort without departing from the physiological and psychological perception of the human body and establish a quantitative comfort evaluation method is a key problem which is tried and solved by students at home and abroad in recent years.

In addition, the control system of the automatic driving vehicle makes corresponding control actions according to different traffic environment states, and slight traffic environment changes can cause different control actions. Purely automotive-based products do not facilitate repeated, quantitative studies in a variety of specific scenarios. The traditional method for controlling the vehicle by adopting a human driver is difficult to realize repeatable fine operation. Therefore, the method for evaluating the comfort level of the driver based on the physiological signals and subjective evaluation and with high precision and repeatability in the experimental process is significant.

Disclosure of Invention

The invention aims to solve the technical problem of providing an automatic driving automobile comfort level test and evaluation method and system based on an automatic driving robot, overcomes the defects of the existing driver comfort level test and evaluation method, and is a driver comfort level test and evaluation method based on the automatic driving robot and electrocardiogram data.

The technical problem to be solved by the invention is realized by the following technical scheme:

an automatic driving automobile comfort level test evaluation method based on an automatic driving robot comprises the following steps:

selecting a plurality of tested drivers, and preparing to adopt an electrocardiosignal physiological test recorder to collect electrocardiosignals;

acquiring electrocardiosignals of a tested driver, and controlling the vehicle to perform preset movement through the automatic driving robot; recording electrocardio data of a tested driver in the whole preset vehicle motion process and vehicle state information controlled by the automatic driving robot;

after finishing the preset vehicle movement, the driver carries out subjective comfort feeling evaluation according to the subjective evaluation table;

according to the electrocardiosignals obtained in the second step, R wave extraction is carried out on the electrocardiosignals of the tested driver, and time domain calculation is carried out on the processing result to obtain time domain indexes SDNN (standard deviation of interval between two R waves of electrocardiogram) and RMSSD (root mean square of interval difference between two adjacent RR) representing heart rate variability; the SDNN is a standard deviation of an interval between two R waves of an electrocardiogram; the RMSSD is the root mean square of the difference value of two adjacent RR intervals.

Carrying out variance analysis according to the physiological data SDNN and the RMSSD obtained in the step four to obtain objective data which have obvious difference and can represent comfort;

and step six, carrying out cross comparison according to the objective data obtained in the step five and the subjective evaluation data obtained in the step three to obtain the optimal vehicle motion state parameter which accords with the comfort level of the driver.

Preferably, in the subjective evaluation form of the third step, the scoring criteria can be directly inputted into the computer through a form stored in the computer or the mobile terminal.

The subjective scoring criteria are as in table 1 below, and after each experiment, the driver performs detailed scoring according to subjective feeling to give subjective comfort evaluation:

TABLE 1 Scoring standards

Preferably, the analysis of the objective data in the third step is obtained by programming calculation, and heart rate variability time domain index SDNN (standard deviation of interval between two R waves of electrocardiogram) and RMSSD (root mean square of difference between two adjacent RR intervals) extracted from the electrocardiographic data are analyzed; and analyzing the change conditions of the SDNN and the RMSSD in combination with the subjective evaluation to obtain corresponding comfortable motion state parameters.

Preferably, the corresponding comfort motion state parameter refers to a maximum longitudinal deceleration, a maximum lateral acceleration parameter in a comfort state.

Preferably, the vehicle state information in the second step includes target acceleration and deceleration, speed and lane change time; the full-automatic driving operation state of the vehicle is simulated through the automatic driving robot in the whole vehicle moving process.

Preferably, the repeated acceleration test precision of the automatic driving automobile is controlled within 0.5m/s 2.

The invention also provides an automatic driving automobile comfort level testing system based on the automatic driving robot, which comprises the following components:

a vehicle having a cockpit;

the automatic driving robot is rigidly and fixedly connected in the vehicle and used for sensing and controlling the vehicle to move according to a set motion state;

the physiological data acquisition equipment is fixed on a driver through an electrocardio sensor and acquires electrocardiosignals through the electrocardio sensor.

Preferably, the autonomous robot further includes:

the pose and motion sensing system of the automatic driving robot is used for sensing the motion state and position of the vehicle and calculating control instruction information by comparing the actual motion state with the target motion state set in advance;

the automatic driving robot control system is used for sending the control instruction information to the automatic driving robot execution system;

and the automatic driving robot executing system is used for controlling the accelerator, the brake and the steering of the vehicle so as to enable the vehicle to move according to the set motion state.

Preferably, the physiological data acquisition device further comprises a data acquisition analysis system; the electrocardiosignals are input into the data acquisition and analysis system and are output for analysis after time synchronization of the electrocardiosignals and the vehicle motion state information output by the automatic driving robot.

Preferably, the physiological data acquisition device is a physiological test recorder, and the test electrodes are arranged by using a II lead method.

Particularly, the information, signals and instruction transmission after the data acquisition can be realized by a known communication connection transmission mode; the system components can be connected by known electrical or mechanical connection means.

The invention discloses a method and a system for testing and evaluating the comfort level of an automatic driving automobile based on an automatic driving robot, which have the following beneficial effects by adopting the technical scheme:

1. the invention discloses an automatic driving automobile comfort level test evaluation method based on an automatic driving robot and a physiological collector, which is used for synchronously and highly accurately collecting subjective and objective data, and performing cross comparison to obtain a comfort level judgment basis of mutual evidence so as to determine the optimal vehicle motion state.

2. The invention relates to an automatic driving automobile comfort level test evaluation system based on an automatic driving robot and a physiological acquisition instrument, which is a vehicle control method based on the automatic driving robot, so that a high-precision repeatable strong simulation automatic driving state is obtained.

3. Compared with the real-time vehicle test directly based on the automatic driving vehicle in the prior art, the method and the system have higher safety, precision and repeatability. The repeated acceleration test precision can be controlled within 0.5m/s 2, and the accuracy and the time cost of real vehicle test are greatly improved. In addition, the analysis and evaluation method for performing cross comparison analysis on objective data and subjective evaluation provides a good objective basis for the comfort evaluation of a driver, and can provide quantitative data support.

Drawings

The present invention will be described in detail below with reference to the accompanying drawings so that the above advantages of the present invention will be more apparent.

FIG. 1 is a schematic diagram of the system of the present invention

FIG. 2 is a schematic diagram of R-wave identification of electrocardiographic data according to the present invention

FIG. 3 is a schematic diagram of the heart rate variability analysis result of the electrocardiographic data

FIG. 4 is a schematic diagram of the recording of the motion status of a vehicle during the testing process of the present invention

FIG. 5 is a schematic diagram showing the subjective and objective comparative analysis results of the present invention

Detailed Description

The system composition diagram of the present invention shown in fig. 1, the subjective evaluation criteria shown in table 1 of the test data shown in fig. 2, fig. 3, fig. 4, and fig. 5, and the present invention are specifically described below.

Example 1

An automatic driving automobile comfort level test evaluation method based on an automatic driving robot and a physiological acquisition instrument comprises the following steps:

selecting a plurality of tested drivers, adopting an electrocardiosignal physiological test recorder as signal acquisition equipment, and arranging test electrodes by using a II-lead method;

step two, when the electrocardiosignals of the tested driver are collected, the automatic driving robot controls the vehicle to carry out preset vehicle motion, and records the electrocardio data and the vehicle state information of the tested driver in the whole experiment process;

after the experiment is completed, the driver carries out subjective feeling evaluation according to the subjective evaluation table;

according to the electrocardiosignals obtained in the second step, R wave extraction is carried out on the electrocardiosignals of the tested driver, and time domain calculation is carried out on the processing result to obtain time domain indexes SDNN and RMSSD representing heart rate variability;

analyzing variance of the physiological data obtained in the step four to obtain objective data which has obvious difference and can represent comfort;

and step six, according to the subjective and objective data obtained in the step five and the step three, cross-comparing to obtain the optimal vehicle motion state parameters which accord with the comfort level of the driver, such as the maximum deceleration and the maximum acceleration which accord with the comfort level of the driver.

Example 2

An automatic driving automobile comfort level test evaluation method based on an automatic driving robot and a physiological acquisition instrument comprises the following steps:

selecting a plurality of tested drivers, adopting an electrocardiosignal physiological test recorder as signal acquisition equipment, and arranging test electrodes by using a II-lead method;

step two, when the electrocardiosignals of the tested driver are collected, the automatic driving robot controls the vehicle to carry out preset vehicle motion, and records the electrocardio data and the vehicle state information of the tested driver in the whole experiment process;

after the experiment is completed, the driver carries out subjective feeling evaluation according to the subjective evaluation table; subjective sensory evaluation criteria are given in table 1 as scoring criteria:

TABLE 1 Scoring standards

According to the electrocardiosignals obtained in the second step, R wave extraction is carried out on the electrocardiosignals of the tested driver, and time domain calculation is carried out on the processing result to obtain time domain indexes SDNN and RMSSD representing heart rate variability;

analyzing variance of the physiological data obtained in the step four to obtain objective data which has obvious difference and can represent comfort;

and step six, according to the subjective and objective data obtained in the step five and the step three, cross-comparing to obtain the optimal vehicle motion state parameters which accord with the comfort level of the driver, such as the maximum deceleration and the maximum acceleration which accord with the comfort level of the driver.

Example 3

Referring to fig. 1, a left block diagram of the system for testing the comfort level of the automatically driven vehicle based on the automatically driven robot and the physiological acquisition instrument is a vehicle without an automatic driving function in the prior art, a driver needs to rapidly change acceleration and speed when the road condition of the vehicle changes, such as braking, sudden stop, lane change and the like, the driver and passengers in the vehicle feel comfortable at different degrees, and the driver often faces danger due to deviation of driving judgment caused by rapid change of the comfort level because of great individual difference. The invention not only designs the automatic driving robot and the operation control system thereof in the existing vehicle, but also further compares the change of the vehicle motion state when the robot automatically drives with the subjective perception of the driver to research the comfort level so as to find the acceleration and the speed of the automatic driving in different vehicle running states in the comfortable state.

The passenger comfort in the invention is a comprehensive experience of interaction of human physiological and psychological factors and a driving state when the motion state of the automatic driving vehicle is changed. When the motion state of the vehicle changes, different degrees of abrupt changes of acceleration and speed are generated, so that various sensory stimuli to passengers are generated, and the comfort degree is changed.

The integral test system consists of a vehicle, an automatic driving robot and a physiological data acquisition instrument, wherein the automatic driving robot is rigidly fixed in the vehicle, and the physiological data acquisition instrument is fixed on a driver through a sensor. The vehicle motion state and position can be sensed through the pose and motion sensing system of the automatic driving robot, and control information is calculated through comparing the actual motion state with the preset target motion state, such as target acceleration and deceleration, speed, lane changing time and the like. The control system sends the control instruction to the automatic driving robot execution system, and the automatic driving robot controls the accelerator, the brake and the steering of the vehicle so as to enable the vehicle to move according to the set motion state. The automatic driving state of the vehicle is simulated through the automatic driving robot in the whole moving process, and a driver does not need to do any operation. The physiological data acquisition instrument acquires electrocardio signals through the sensor, and the signals and the vehicle motion state information output by the automatic driving robot are output after time synchronization for analysis.

The analysis method of the system mainly comprises a main data analysis part and an objective data analysis part, and is developed aiming at N tested drivers with corresponding characteristics according to requirements. After each experiment, the driver was scored in detail according to subjective feelings, as shown in table 1, giving a subjective comfort assessment. The analysis of the objective data is obtained by programming calculation, and mainly analyzes a heart rate variability time domain index SDNN (standard deviation of interval between two R waves of electrocardiogram) and an RMSSD (root mean square of difference between two adjacent RR intervals) extracted from the electrocardiographic data, as shown in fig. 1 and 2. By analyzing the change conditions of the SDNN and the RMSSD in combination with subjective evaluation, corresponding comfort motion state parameters, such as maximum longitudinal deceleration, maximum lateral acceleration, etc. in a comfort state, are obtained, as shown in fig. 4 and 5.

The repeated acceleration test precision of the automatic driving automobile can be controlled within 0.5m/s 2.

And (4) data conclusion:

the R-wave identification of the electrocardiographic data shown in fig. 2 yields: the designed experimental process and experimental environment can ensure that a stable electrocardiogram which can be used for data analysis can be obtained in the test process.

The result of the programmed analysis of the electrocardiographic data shown in fig. 3 is based on the stable electrocardiographic data shown in fig. 2, and the RMSSD and SDNN values in the experiment can be calculated by the programmed program. The program is written according to the C language, and the operation environment is in matlab to obtain the data of the diagram. Different sets of data are available for analysis for different states.

The data records of the vehicle motion states (longitudinal acceleration, speed, position) shown in fig. 4 are represented by three curves from top to bottom: the speed of the test vehicle, the relative distance of the test vehicle from a position reference point (a randomly selected position point in the test process, which is used as a reference point to record the position of the vehicle), and the acceleration of the test vehicle. The method has the advantages that the speed, the position and the deceleration of the vehicle can be accurately recorded in the test process, SDNN RMSSD changes before and after the deceleration can be accurately corresponding in time on the basis, and the problem that corresponding external stimuli (deceleration) cannot be found due to the changes of the RMSSD and the SDNN is solved. Based on the above results, objective and subjective analysis on comfort can be finally conducted.

The results of the evaluation analysis shown in fig. 5, the acceleration of fig. 5 being the longitudinal acceleration of fig. 4, show: the specific positive relation of the acceleration variability time domain index is as follows: as the deceleration increases (from-2 to-6), the heart rate variability index decreases, with a negative correlation.

The acceleration and the heart rate are analyzed to be consistent with the subjective score slope in the driving process, and the following results are obtained: the acceleration changes, the heart rate variability changes, and the subjective score changes are strongly, but not simply linearly related. The slope is therefore not uniform throughout the change, the slopes of different deceleration intervals are not uniform, but the overall trend is uniform.

The slope change node changes in parameters or time or forecast values, and corresponding score analysis is carried out to obtain: it can be seen from the graph that the slope node changes significantly when the acceleration is equal to-4, with a corresponding subjective score of 4.

In summary, the automatic driving automobile comfort level test evaluation system based on the automatic driving robot has the following advantages:

the vehicle motion state and position are sensed through the pose and motion sensing system of the automatic driving robot, the actual motion state is compared with the preset target motion state, such as target acceleration and deceleration, speed, lane changing time and the like, the automatic driving state of the vehicle is simulated through the automatic driving robot in the whole motion process, and a driver does not need to do any operation. The vehicle control method based on the automatic driving robot obtains a high-precision and strong-repeatability simulated automatic driving state; the method overcomes the defects of the existing driver comfort test and evaluation method.

The method aims at different working conditions, such as different longitudinal acceleration and deceleration, braking, lane changing and other working conditions, the automatic driving robot is used for accurately controlling the motion state and the pose of the vehicle, electrocardiosignals of a driver in a specific vehicle motion state are extracted, and data are analyzed by combining subjective evaluation, so that the method for testing and evaluating the comfort level of the driver of the automatic driving vehicle based on the electrocardiosignals and the vehicle motion state is formed.

The test evaluation method has high precision for controlling the motion state of the vehicle, has strong repeatability and usability, and has good compatibility to different working conditions. In addition, the method can cross-compare the electrocardiogram data with subjective evaluation, provides a better objective basis for the evaluation of the comfort level of the driver, and is beneficial to research on the control strategy of the automatic driving vehicle by a vehicle enterprise and detection mechanism so as to improve the riding comfort of the driver.