Calculation method for driver takeover interaction performance in non-driving posture

Abstract

The invention discloses a method for calculating driver takeover interaction performance in a non-driving posture, and relates to the field of traffic safety. Virtual simulation software is utilized to construct an experiment scene to carry out man-machine co-driving vehicle driver takeover experiments, different seat backrest angles are set, and multiple non-driving tasks are arranged for a driver; collecting data of a plurality of drivers steering to take over vehicle manipulation from various non-driving tasks under different seat backrest angles, and calculating weights of various indexes under each seat backrest angle by using an entropy method; utilizing a fuzzy comprehensive evaluation method to obtain a fuzzy comprehensive evaluation matrix of the pipe connection timeliness and the pipe connection stability under each seat backrest angle; obtaining a driver takeover interaction performance evaluation score in a non-driving posture; through three dimensions of taking-over timeliness, taking-over stability and driving comfort, taking-over effectiveness and driving safety of a driver under different seat backrest angles are evaluated more comprehensively, and the evaluation is more reasonable.

Description

Technical Field

[0001] The invention relates to the field of traffic safety, and in particular to a method for calculating driver takeover interaction performance in a non-driving posture. Background Art

[0002] In the human-machine co-driving stage, the driver and the car's automatic driving system jointly undertake the driving task. In the automatic driving mode, the driver is allowed to engage in non-driving related tasks, such as using a mobile phone, eating, etc. Some drivers will adjust the seat back to a more comfortable non-driving posture to engage in non-driving related tasks. However, when encountering a situation that the automatic driving system cannot handle, the driver is still required to take over the vehicle and respond immediately to the takeover event to avoid traffic accidents. In this case, if the driver's takeover ability is weak, and he fails to take over the vehicle in time or fails to operate the vehicle in time after taking over the vehicle, a traffic accident is very likely to occur.

[0003] Most existing technologies focus on seat comfort evaluation, improving the driver's driving convenience and optimizing the driving experience, providing guidance and basis for the optimal design of seat posture adjustment mode. However, there is currently a lack of quantitative analysis of the driver's takeover interaction performance at different seat back tilt angles in autonomous driving mode, and a lack of qualitative analysis on how to ensure the effectiveness of takeover and driving comfort.

[0004] Based on this, the present invention aims to provide a method for calculating the driver's takeover interaction performance in a non-driving posture, so as to achieve a qualitative analysis of the takeover under different seat back tilt angles. Summary of the invention

[0005] In order to achieve qualitative analysis of takeover under different seat back tilt angles, the purpose of the present invention is to provide a method for calculating the driver's takeover interaction performance in a non-driving posture. The method is based on the driver's takeover data and vehicle operation data, uses the entropy method to evaluate the weights of various data indicators, and uses the fuzzy comprehensive evaluation method to comprehensively evaluate the driver's takeover ability, and then uses the obtained scores and weights to derive a comprehensive score of the driver's takeover interaction performance under different seat back angles.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for calculating driver takeover interaction performance in a non-driving posture comprises the following steps:

[0008] (1) Use virtual simulation software to build an experimental scenario to conduct a human-machine co-driving vehicle driver takeover experiment, set different seat back angles, and assign multiple non-driving tasks to the driver;

[0009] (2) Collect data on multiple drivers switching from non-driving tasks to taking over vehicle control at different seat back angles. The data includes three types of indicators: timeliness of takeover, control stability, and seat comfort.

[0010] (3) Processing the data in step (2), establishing a matrix of the index data of takeover timeliness, handling stability, and seat comfort of multiple drivers at the same seat back angle, and calculating the weight of each index at each seat back angle using an entropy method;

[0011] (4) Establishing a set of comprehensive evaluation indicators for the driver's takeover timeliness and takeover stability at different seat back angles, using all the data collected in step (2) to classify the comprehensive evaluation indicators of the driver's takeover timeliness and takeover stability, and using the fuzzy comprehensive evaluation method to obtain a fuzzy comprehensive evaluation matrix for the takeover timeliness and takeover stability at each seat back angle;

[0012] (5) Based on the weights of various indicators obtained in step (3) and the fuzzy comprehensive evaluation matrix of takeover timeliness and takeover stability obtained in step (4), the driver's takeover interaction performance evaluation scores in non-driving postures at different seat back angles are calculated respectively, and the driver's takeover interaction performance evaluation scores in non-driving postures are obtained.

[0013] In the present invention, the index data of takeover timeliness include takeover reaction time, first brake pedal pressing time and first road looking time; the index data of takeover stability include brake pedal force, steering wheel angle, heart rate and RR interval; the index data of seat comfort is obtained by conducting a questionnaire survey on drivers.

[0014] In the present invention, the specific process of calculating the weights of various indicators at each seat back angle using the entropy method is as follows:

[0015] The driver's takeover interaction performance raw data matrix:

[0016]

[0017] Takeover timeliness raw data matrix: Take over the original data matrix of stability: In the formula, a i1 ——Indicator data of the takeover reaction time of the i-th driver;

[0018] a i2 ——The first brake pedal pressing time index data of the i-th driver;

[0019] a i3 ——The first time the i-th driver looks at the road;

[0020] a i4 ——Brake pedal force index data of the i-th driver;

[0021] a i5 ——The steering wheel angle index data of the i-th driver;

[0022] a i6 ——Heart rate index data of the i-th driver;

[0023] a i7 ——Driving comfort index data of the i-th driver;

[0024] Positive indicator normalization: Negative indicator normalization: After standardization, the following new driver takeover interaction performance data matrix A is obtained: 1 :

[0025]

[0026] Takeover timeliness data matrix: Takeover stability data matrix: The numerical weight of the driver takeover interaction performance index is calculated by the following formula:

[0027] In the formula, B ij ——The numerical weight of the jth indicator of the ith driver; the numerical weight of the driver’s takeover timeliness indicator is calculated by the following formula:

[0028] In the formula, ——The numerical weight of the jth indicator of the ith driver; the numerical weight of the driver's takeover stability indicator is calculated by the following formula:

[0029] In the formula, ——The numerical weight of the jth indicator of the ith driver; the k value is calculated by the following formula:

[0030]

[0031] The entropy value of the driver takeover interaction performance index e is calculated by the following formula j :

[0032] In the formula, e j ——The entropy value of the jth indicator;

[0033] The entropy value of the driver takeover timeliness index is calculated by the following formula

[0034] The entropy value of the driver takeover stability index is calculated by the following formula

[0035] The information entropy redundancy d of the driver takeover interaction performance indicator is calculated by the following formula j :d j =1-e j (j=1,2,...,7)

[0036] The information entropy redundancy of the driver takeover timeliness index is calculated by the following formula

[0037] The information entropy redundancy of the driver takeover stability index is calculated by the following formula

[0038] The weights of each data indicator of driver takeover interaction performance w are calculated by the following formula j :

[0039] The weights of various data indicators for driver takeover timeliness are calculated by the following formula

[0040] The weights of various data indicators of driver takeover stability are calculated by the following formula

[0041] Therefore, the indicator weight matrix of the driver takeover interaction performance under different seat back angles is: W = [w1 w2w3 w4 w5 w6 w7]

[0042] Where, w1 is the weight of the driver’s takeover reaction time indicator for interactive performance;

[0043] w2——the weight of the driver’s first brake pedal stepping time indicator for taking over the interaction performance;

[0044] w3——the weight of the first time the driver looks at the road when taking over the interaction performance;

[0045] w4——the weight of the brake pedal force index of the driver’s takeover interaction performance;

[0046] w5——the weight of the steering wheel angle indicator of the driver’s takeover interaction performance;

[0047] w6——the weight of the driver’s heart rate indicator for driver takeover interaction performance;

[0048] w7——the weight of the driver’s driving comfort index of the driver’s takeover interaction performance;

[0049] The index weight matrix of driver takeover timeliness under different seat back angles is obtained as follows:

[0050]

[0051] In the formula, ——The weight of the driver’s takeover timeliness and response time indicator;

[0052] ——The weight of the first brake pedal stepping time indicator for the driver to take over promptly;

[0053] ——The weight of the driver’s first road-watching time indicator;

[0054] The index weight matrix of driver takeover stability under different seat back angles is obtained as follows:

[0055]

[0056] In the formula, - the weight of the driver takeover stability brake pedal force indicator;

[0057] ——The weight of the driver’s steering wheel angle indicator for driver takeover stability;

[0058] ——The weight of the driver’s heart rate indicator for driver takeover stability.

[0059] In the present invention, the calculation process of the fuzzy comprehensive evaluation matrix of takeover timeliness and takeover stability is as follows:

[0060] Establish a comprehensive evaluation factor set for the driver's timely takeover at different seat back angles:

[0061] X 1 ={takeover reaction time, first brake pedal application time, first road gaze time}

[0062] Establish a comprehensive evaluation factor set for driver takeover stability at different seat back angles:

[0063] X 2 ={brake pedal force, steering wheel angle, driver's heart rate}

[0064] Construct the following indicator evaluation set:

[0065] Y=[y1 y2 y3 y4 y5]

[0066] In the formula, y1——the fuzzy level of the indicator is excellent;

[0067] y2——the fuzzy level of the indicator is good;

[0068] y3——the indicator fuzziness level is medium;

[0069] y4——The fuzzy level of the indicator is poor;

[0070] y5——the fuzziness level of the indicator is poor;

[0071] Based on the data collected from the experiment, determine the maximum value a of all i-th indicators imin , minimum value a imax , divided into five levels from small to large: excellent, good, medium, poor, and bad. Among them, the excellent level (a imin , b i1 ), good grade (b i1 , b i2 ), medium level (b i2 , b i3 ), poor grade (b i3 , b i4 ), poor grade (b i4 , a imax ),

[0072]

[0073] The membership degree of the takeover timeliness and takeover stability indicators at their fuzzy levels is calculated by the following formula:

[0074] Blur level is excellent:

[0075] Blur level is good:

[0076] Medium blur level:

[0077] Blur level is poor:

[0078] Blur level is poor:

[0079] In the formula, a ij ——The jth data value of the ith indicator;

[0080] ——The degree of membership of the i-th indicator of driver takeover timeliness at different fuzzy levels;

[0081] ——the degree of membership of the i-th indicator of driver takeover stability at different fuzzy levels;

[0082] The fuzzy comprehensive evaluation matrix of driver takeover timeliness under different seat back angles is obtained as follows:

[0083]

[0084] The fuzzy comprehensive evaluation matrix of driver takeover stability under different seat back angles is obtained as follows:

[0085]

[0086] In the present invention, the calculation process of the driver takeover interaction performance evaluation score in the non-driving posture is as follows:

[0087] The fuzzy vector R of the driver's takeover timeliness at different seat back angles is calculated by the following formula 1 :

[0088]

[0089] The driver's takeover stability fuzzy vector R at different seat back angles is calculated by the following formula 2 :

[0090]

[0091] In the following the score matrix is ​​set:

[0092] F = {excellent, good, average, poor, bad} = [1007550250]

[0093] The driver takeover timeliness score t at different seat back angles is obtained by the following formula 1 :

[0094]

[0095] The driver takeover timeliness score t at different seat back angles is obtained by the following formula 2 :

[0096]

[0097] The questionnaire results show that the driver's driving comfort score at different seat back angles is t 3 ;

[0098] Finally, the driver takeover interaction performance evaluation score T in non-driving posture is obtained by the following formula:

[0099] T=t 1 ×(w1+w2+w3)+t 2 ×(w4+w5+w6)+t 3 ×w7.

[0100] Compared with the prior art, the present invention has the following beneficial effects:

[0101] (1) The data collected by the present invention are divided into three dimensions, namely, takeover timeliness, takeover stability and driving comfort, and the driver's takeover behavior is scored in the three dimensions respectively, so as to conduct a more comprehensive evaluation of the driver's takeover effectiveness and driving safety at different seat back angles.

[0102] (2) The present invention combines the entropy method and the fuzzy evaluation method to comprehensively evaluate the driver's takeover timeliness, takeover stability and driving comfort, and obtains an evaluation score. It quantifies the takeover effectiveness and driving comfort at different seat back angles, and more intuitively evaluates the driver's takeover behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0104] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0105] The invention discloses a method for calculating the driver takeover interaction performance in a non-driving posture.

[0106] The method is applied to an autonomous driving vehicle in autonomous driving mode. Through steps 1 to 5, the vehicle operation data, eye movement data and physiological data of the driver at different seat back tilt angles are collected. The weight of each data indicator is calculated using the entropy method. The fuzzy comprehensive evaluation method is used to comprehensively evaluate the timeliness and stability of the driver's takeover. The obtained scores and weights are then used to calculate the comprehensive scores of the driver's takeover interaction performance at different seat back angles. The flowchart is shown in FIG. Figure 1 shown.

[0107] Step 1: Use virtual simulation software to construct an experimental scenario and conduct a driver takeover experiment for a human-machine co-driving vehicle. In the experiment, the driver adjusted the seat back to three angles: 100°, 110°, and 120°, and engaged in three non-driving related tasks: playing with the phone, eating, and sleeping.

[0108] Step 2: Use a driving simulator to collect the vehicle control data of the driver taking over, including brake pedal force, first brake pedal pressing time, takeover reaction time and steering wheel angle; use a physiological recorder to collect the driver's physiological index data under different seat backs and different non-driving tasks; use an eye tracker to collect the driver's eye movement index data, including the first time to look at the road; use a questionnaire to collect the driver's subjective rating of driving comfort at different seat back angles. After collection, the data is divided into three types of evaluation data: takeover timeliness, handling stability and seat comfort. Specifically, takeover timeliness includes takeover reaction time, first brake pedal pressing time and first road looking time indicators, takeover stability includes brake pedal force, steering wheel angle and heart rate, and seat comfort is subjectively evaluated by the driver's questionnaire.

[0109] Step 3: Based on the data from step 2, a comprehensive evaluation is conducted on the driver's takeover timeliness, takeover stability, and driving comfort at different seat back angles. First, the entropy method is used to calculate the weights of various data indicators, and the data is standardized using the following formula:

[0110] The driver's takeover interaction performance raw data matrix:

[0111]

[0112] Takeover timeliness raw data matrix:

[0113] Take over the original data matrix of stability:

[0114] In the formula, a i1 ——Indicator data of the takeover reaction time of the i-th driver;

[0115] a i2 ——The first brake pedal pressing time index data of the i-th driver;

[0116] a i3 ——The first time the i-th driver looks at the road;

[0117] a i4 ——Brake pedal force index data of the i-th driver;

[0118] a i5 ——The steering wheel angle index data of the i-th driver;

[0119] a i6 ——Heart rate index data of the i-th driver;

[0120] a i7 ——Driving comfort index data of the i-th driver;

[0121] Positive indicator normalization: Negative indicator normalization: After standardization, the following new driver takeover interaction performance data matrix A is obtained: 1 :

[0122]

[0123] Takeover timeliness data matrix: Takeover stability data matrix: The numerical weight of the driver takeover interaction performance index is calculated by the following formula:

[0124]

[0125] In the formula, B ij ——The numerical weight of the jth indicator of the ith driver;

[0126] The numerical weight of the driver takeover timeliness index is calculated by the following formula:

[0127]

[0128] In the formula, ——The numerical weight of the jth indicator of the ith driver;

[0129] The numerical weight of the driver takeover stability index is calculated by the following formula:

[0130]

[0131] In the formula, ——The numerical weight of the jth indicator of the ith driver;

[0132] The k value is calculated by the following formula:

[0133]

[0134] The entropy value of the driver takeover interaction performance index e is calculated by the following formula j :

[0135]

[0136] In the formula, e j ——The entropy value of the jth indicator;

[0137] The entropy value of the driver takeover timeliness index is calculated by the following formula

[0138] The entropy value of the driver takeover stability index is calculated by the following formula

[0139] The information entropy redundancy d of the driver takeover interaction performance indicator is calculated by the following formula j :d j =1-e j (j=1,2,...,7)

[0140] The information entropy redundancy of the driver takeover timeliness index is calculated by the following formula

[0141] The information entropy redundancy of the driver takeover stability index is calculated by the following formula

[0142] The weights of each data indicator of driver takeover interaction performance w are calculated by the following formula j :

[0143]

[0144] The weights of various data indicators for driver takeover timeliness are calculated by the following formula

[0145]

[0146] The weights of various data indicators of driver takeover stability are calculated by the following formula

[0147]

[0148] Therefore, the indicator weight matrix of the driver takeover interaction performance under different seat back angles is: W = [w1 w2w3 w4 w5 w6 w7]

[0149] Where, w1 is the weight of the driver’s takeover reaction time indicator for interactive performance;

[0150] w2——the weight of the driver’s first brake pedal stepping time indicator for taking over the interaction performance;

[0151] w3——the weight of the first time the driver looks at the road when taking over the interaction performance;

[0152] w4——the weight of the brake pedal force index of the driver’s takeover interaction performance;

[0153] w5——the weight of the steering wheel angle indicator of the driver’s takeover interaction performance;

[0154] w6——the weight of the driver’s heart rate indicator for driver takeover interaction performance;

[0155] w7——the weight of the driver’s driving comfort index of the driver’s takeover interaction performance;

[0156] The index weight matrix of driver takeover timeliness under different seat back angles is obtained as follows:

[0157]

[0158] In the formula, ——The weight of the driver’s takeover timeliness and response time indicator;

[0159] ——The weight of the first brake pedal stepping time indicator for the driver to take over promptly;

[0160] ——The weight of the driver’s first road-watching time indicator;

[0161] The index weight matrix of driver takeover stability under different seat back angles is obtained as follows:

[0162]

[0163] In the formula, - the weight of the driver takeover stability brake pedal force indicator;

[0164] ——The weight of the driver’s steering wheel angle indicator for driver takeover stability;

[0165] ——The weight of the driver's heart rate indicator for driver takeover stability;

[0166] Step 4: Based on the weights of each indicator calculated in step 3, the fuzzy comprehensive evaluation method is used to obtain the fuzzy comprehensive evaluation matrix of takeover timeliness and takeover stability.

[0167] Establish a comprehensive evaluation factor set for the driver's timely takeover at different seat back angles:

[0168] X 1 ={takeover reaction time, first brake pedal application time, first road gaze time}

[0169] Establish a comprehensive evaluation factor set for driver takeover stability at different seat back angles:

[0170] X 2 ={brake pedal force, steering wheel angle, driver's heart rate}

[0171] Construct the following indicator evaluation set:

[0172] Y=[y1 y2 y3 y4 y5]

[0173] In the formula, y1——the fuzzy level of the indicator is excellent;

[0174] y2——the fuzzy level of the indicator is good;

[0175] y3——the indicator fuzziness level is medium;

[0176] y4——The fuzzy level of the indicator is poor;

[0177] y5——the fuzziness level of the indicator is poor;

[0178] The data collected in the experiment are divided into five levels according to the data size. Specifically, the indicator data are divided into five levels from small to large: excellent, good, medium, poor, and bad. The following evaluation indicator fuzzy level table is constructed:

[0179]

[0180] In the formula, a imax ——The maximum value of the i-th indicator;

[0181] a imin ——The minimum value of the i-th indicator;

[0182] The value b of each level of the i-th indicator is calculated by the following formula: ij :

[0183]

[0184] The membership degree of the takeover timeliness and takeover stability indicators at their fuzzy levels is calculated by the following formula:

[0185] Blur level is excellent:

[0186] Blur level is good:

[0187] Medium blur level:

[0188] Blur level is poor:

[0189] Blur level is poor:

[0190] In the formula, a ij ——The jth data value of the ith indicator;

[0191] ——The degree of membership of the i-th indicator of driver takeover timeliness at different fuzzy levels;

[0192] ——the degree of membership of the i-th indicator of driver takeover stability at different fuzzy levels;

[0193] The fuzzy comprehensive evaluation matrix of driver takeover timeliness under different seat back angles is obtained as follows:

[0194]

[0195] The fuzzy comprehensive evaluation matrix of driver takeover stability under different seat back angles is obtained as follows:

[0196]

[0197] Step 5: According to the indicator weight matrix W calculated in step 3 1 , W 2 And the fuzzy comprehensive evaluation matrix Z calculated in step 4 1 , Z 2 , establish a comprehensive evaluation, and derive scores for the driver's takeover timeliness and takeover stability at different seat back angles.

[0198] The fuzzy vector R of the driver's takeover timeliness at different seat back angles is calculated by the following formula 1 :

[0199]

[0200] The driver's takeover stability fuzzy vector R at different seat back angles is calculated by the following formula 2 :

[0201]

[0202] In this paper, the scoring matrix is ​​set as follows:

[0203] F = {excellent, good, average, poor, bad} = [100 75 50 25 0]

[0204] The driver takeover timeliness score t at different seat back angles is obtained by the following formula 1 :

[0205]

[0206] The driver takeover timeliness score t at different seat back angles is obtained by the following formula 2 :

[0207]

[0208] The questionnaire results show that the driver's driving comfort score at different seat back angles is t 3 ;

[0209] Finally, the driver takeover interaction performance evaluation score T in non-driving posture is obtained by the following formula:

[0210] T=t 1 ×(w1+w2+w3)+t 2 ×(w4+w5+w6)+t 3 ×w7.

[0211] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for calculating the driver takeover interaction performance in a non-driving posture, characterized in that: The following steps are involved: (1) Use virtual simulation software to build an experimental scenario to conduct a human-machine co-driving vehicle driver takeover experiment, set different seat back angles, and assign multiple non-driving tasks to the driver; (2) Collect data on multiple drivers switching from non-driving tasks to taking over vehicle control at different seat back angles. The data includes three types of indicators: timeliness of takeover, control stability, and seat comfort. (3) Processing the data in step (2), establishing a matrix of the index data of takeover timeliness, handling stability, and seat comfort of multiple drivers at the same seat back angle, and calculating the weight of each index at each seat back angle using an entropy method; (4) Establishing a set of comprehensive evaluation indicators for the driver's takeover timeliness and takeover stability at different seat back angles, using all the data collected in step (2) to classify the comprehensive evaluation indicators of the driver's takeover timeliness and takeover stability, and using the fuzzy comprehensive evaluation method to obtain a fuzzy comprehensive evaluation matrix for the takeover timeliness and takeover stability at each seat back angle; (5) Based on the weights of various indicators obtained in step (3) and the fuzzy comprehensive evaluation matrix of takeover timeliness and takeover stability obtained in step (4), the driver's takeover interaction performance evaluation scores in non-driving postures at different seat back angles are calculated respectively, and the driver's takeover interaction performance evaluation scores in non-driving postures are obtained.

2. The method for calculating the driver takeover interaction performance in a non-driving posture according to claim 1, characterized in that: The index data of takeover timeliness include takeover reaction time, first brake pedal pressing time and first road gaze time; the index data of takeover stability include brake pedal force, steering wheel angle, heart rate and RR interval; the index data of seat comfort are obtained by conducting a questionnaire survey on drivers.

3. The method for calculating the driver takeover interaction performance in a non-driving posture according to claim 2, characterized in that: The specific process of using the entropy method to calculate the weights of various indicators at each seat back angle is as follows: The driver's takeover interaction performance raw data matrix: Takeover timeliness raw data matrix: Take over the original data matrix of stability: In the formula, a i1 ——Indicator data of the takeover reaction time of the i-th driver; a i2 ——The first brake pedal pressing time index data of the i-th driver; a i3 ——The first time the i-th driver looks at the road; a i4 ——Brake pedal force index data of the i-th driver; a i5 ——The steering wheel angle index data of the i-th driver; a i6 ——Heart rate index data of the i-th driver; a i7 ——Driving comfort index data of the i-th driver; Positive indicator normalization: Negative indicator normalization: After standardization, the following new driver takeover interaction performance data matrix A is obtained: 1 : Takeover timeliness data matrix: Takeover stability data matrix: The numerical weight of the driver takeover interaction performance index is calculated by the following formula: In the formula, B ij ——The numerical weight of the jth indicator of the ith driver; the numerical weight of the driver’s takeover timeliness indicator is calculated by the following formula: In the formula, ——The numerical weight of the jth indicator of the ith driver; the numerical weight of the driver's takeover stability indicator is calculated by the following formula: In the formula, ——The numerical weight of the jth indicator of the ith driver; the k value is calculated by the following formula: The entropy value of the driver takeover interaction performance index e is calculated by the following formula j : In the formula, e j ——The entropy value of the jth indicator; The entropy value of the driver takeover timeliness index is calculated by the following formula The entropy value of the driver takeover stability index is calculated by the following formula The information entropy redundancy d of the driver takeover interaction performance indicator is calculated by the following formula j :d j =1-e j (j=1,2,...,7) The information entropy redundancy of the driver takeover timeliness index is calculated by the following formula The information entropy redundancy of the driver takeover stability index is calculated by the following formula The weights of each data indicator of driver takeover interaction performance w are calculated by the following formula j : The weights of various data indicators for driver takeover timeliness are calculated by the following formula The weights of various data indicators of driver takeover stability are calculated by the following formula Therefore, the indicator weight matrix of the driver takeover interaction performance under different seat back angles is: W = [w1w2w3w4w5w6w7] Where, w1 is the weight of the driver’s takeover reaction time indicator for interactive performance; w2——the weight of the driver’s first brake pedal stepping time indicator for taking over the interaction performance; w3——the weight of the first time the driver looks at the road when taking over the interaction performance; w4——the weight of the brake pedal force index of the driver’s takeover interaction performance; w5——the weight of the steering wheel angle indicator of the driver’s takeover interaction performance; w6——the weight of the driver’s heart rate indicator for driver takeover interaction performance; w7——the weight of the driver’s driving comfort index of the driver’s takeover interaction performance; The index weight matrix of driver takeover timeliness under different seat back angles is obtained as follows: In the formula, ——The weight of the driver’s takeover timeliness and response time indicator; ——The weight of the first brake pedal stepping time indicator for the driver to take over promptly; ——The weight of the driver’s first road-watching time indicator; The index weight matrix of driver takeover stability under different seat back angles is obtained as follows: In the formula, - the weight of the driver takeover stability brake pedal force indicator; ——The weight of the driver’s steering wheel angle indicator for driver takeover stability; ——The weight of the driver’s heart rate indicator for driver takeover stability.

4. The method for calculating the driver takeover interaction performance in a non-driving posture according to claim 3, characterized in that: The calculation process of the fuzzy comprehensive evaluation matrix of takeover timeliness and takeover stability is as follows: Establish a comprehensive evaluation factor set for the driver's timely takeover at different seat back angles: X 1 ={takeover reaction time, first brake pedal application time, first road gaze time} Establish a comprehensive evaluation factor set for driver takeover stability at different seat back angles: X 2 ={brake pedal force, steering wheel angle, driver's heart rate} Construct the following indicator evaluation set: Y=[y1y2y3y4y5] In the formula, y1——the fuzzy level of the indicator is excellent; y2——the fuzzy level of the indicator is good; y3——the indicator fuzziness level is medium; y4——The fuzzy level of the indicator is poor; y5——the fuzziness level of the indicator is poor; Based on the data collected from the experiment, determine the maximum value a of all i-th indicators imin , minimum value a imax , divided into five levels from small to large: excellent, good, medium, poor, and bad. Among them, the excellent level (a imin , b i1 ), good grade (b i1 , b i2 ), medium level (b i2 , b i3 ), poor grade (b i3 , b i4 ), poor grade (b i4 , a imax ), The membership degree of the takeover timeliness and takeover stability indicators at their fuzzy levels is calculated by the following formula: Blur level is excellent: Blur level is good: Medium blur level: Blur level is poor: Blur level is poor: In the formula, a ij ——The jth data value of the ith indicator; ——The degree of membership of the i-th indicator of driver takeover timeliness at different fuzzy levels; ——the degree of membership of the i-th indicator of driver takeover stability at different fuzzy levels; The fuzzy comprehensive evaluation matrix of driver takeover timeliness under different seat back angles is obtained as follows: The fuzzy comprehensive evaluation matrix of driver takeover stability under different seat back angles is obtained as follows:

5. The method for calculating the driver takeover interaction performance in a non-driving posture according to claim 4, characterized in that: The calculation process of the driver takeover interaction performance evaluation score in non-driving posture is as follows: The fuzzy vector R of the driver's takeover timeliness at different seat back angles is calculated by the following formula 1 : The driver's takeover stability fuzzy vector R at different seat back angles is calculated by the following formula 2 : In the following the score matrix is ​​set: F = {excellent, good, average, poor, bad} = [1007550250] The driver takeover timeliness score t at different seat back angles is obtained by the following formula 1 : The driver takeover timeliness score t at different seat back angles is obtained by the following formula 2 : The questionnaire results show that the driver's driving comfort score at different seat back angles is t 3 ; Finally, the driver takeover interaction performance evaluation score T in non-driving posture is obtained by the following formula: T=t 1 ×(w1+w2+w3)+t 2 ×(w4+w5+w6)+t 3 ×w7。

Images