Electric bicycle rider reaction time prediction method and system

Abstract

The application discloses an electric bicycle rider reaction time prediction method and system, which comprises the following steps: designing an electric bicycle reaction time test scheme, carrying out a real vehicle test on a closed road section and recording a video, collecting data of an electric bicycle rider wearing a helmet, riding speed, riding steering and reaction time according to the video; designing a questionnaire to collect variables of the electric bicycle reaction time of the rider during riding; further screening the influence variables, and converting the classification variables into virtual variables for regression analysis; establishing an electric bicycle reaction time linear regression model with speed and virtual variables as independent variables; inputting the obtained variables into the electric bicycle reaction time linear regression model to obtain the reaction time. The application can provide suggestions for the helmet types of people with different perceptual abilities, and provide a theoretical basis for the calculation of the electric bicycle parking sight distance and conflict time.

Description

Technical Field

[0001] This invention relates to the field of electric bicycle and traffic safety analysis technology, and in particular to a method and system for predicting the reaction time of electric bicycle riders. Background Technology

[0002] The number of electric bicycles is increasing rapidly, and the number of traffic accidents involving electric bicycles remains high. Therefore, it is necessary to understand the cognitive reaction process of riders. The cognitive reaction process is mainly divided into four stages: perception, memory, processing, and reaction (control). Reaction time is an important indicator in the cognitive reaction process.

[0003] Reaction time is the time it takes for a driver to react to a sensory stimulus, and it is primarily used to calculate safe distances. Research on factors influencing driver reaction time is relatively mature, including age, gender, vehicle speed, weather, driving experience, distracted driving, traffic sign information content, traffic landscape, and specific traffic scenarios. However, most existing research is based on motor vehicle driving simulators and real-vehicle experiments, and does not consider the impact of electric bicycles and their helmets. Summary of the Invention

[0004] The technical problem to be solved by this invention is to propose a method and system for predicting the reaction time of electric bicycle riders. This method and system can comprehensively consider the influence of factors such as the rider's helmet, riding speed, steering, physiological characteristics, behavioral load and driving expectations on the rider's reaction time. Furthermore, it constructs a rider reaction time prediction model, providing a theoretical basis for calculating the line of sight and collision distance of electric bicycles, which is of great significance for ensuring the safety of electric bicycles.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] The present invention proposes a method for predicting the reaction time of electric bicycle riders, including...

[0007] S1. Design a test plan for the reaction time of electric bicycles, conduct real-vehicle tests on closed road sections and record videos, and collect data on the rider's helmet wearing, riding speed, riding steering and reaction time based on the video.

[0008] S2. Through relevant literature review, factors that may affect the reaction time of electric bicycles include rider's physiological attributes, riding experience, riding load, and riding expectations; therefore, a questionnaire was designed to collect data on the physiological characteristics, riding load, and riding expectations of the riders participating in the experiment.

[0009] S3. Based on steps S1 and S2, further screen the influencing variables and convert the categorical variables into dummy variables for regression analysis.

[0010] S4. Establish a linear regression model of electric bicycle reaction time using speed and dummy variables as independent variables.

[0011] S5. Input the obtained variables into the linear regression model of electric bicycle reaction time to predict the reaction time.

[0012] Furthermore, in step S1, the collected data includes the following:

[0013] The rider rides under different types of helmets and turning conditions. When an obstacle is detected, the rider brakes to slow down until coming to a stop, and this process is recorded by a video recording device. By playing back the video recording, the time from when the obstacle is detected to when the brake is first applied is recorded as the rider's reaction time. The distance of the road before the obstacle is detected is measured, and the time taken to travel that distance is read from the video to calculate the riding speed.

[0014] Among them, the electric bicycle rider's reaction time is the selected reaction time, which includes the time it takes for the rider's finger to move to the brake.

[0015] Furthermore, in step S1, the helmet worn by the cyclist includes a full-face helmet, a 3 / 4 half helmet, a 1 / 2 half helmet, and not wearing a helmet; the cycling direction includes left turn and right turn.

[0016] Furthermore, in step S2, the variables collected include the following:

[0017] Variables affecting the reaction time of electric bicycles include, but are not limited to, physiological attributes, riding experience, riding load, and riding expectations.

[0018] Physiological attributes include, but are not limited to, age, gender, and vision; cycling experience includes, but is not limited to, the number of days a week the rider has ridden an electric bicycle; cycling load includes, but is not limited to, visual load, i.e., whether the rider believes the helmet will affect vision while riding; cycling expectations include, but are not limited to, safety expectations, which refer to the rider's level of concern for riding safety. The more concerned the rider is about riding safety, the higher their safety expectations will be.

[0019] The questionnaire uses a 5-point Likert scale, with each question having 5 options, such as "What are your safety expectations for electric bicycles? Very high, higher, average, lower, very low." Questions include age, gender, vision, driving experience, number of days a week spent riding an electric bicycle, visual load, behavioral load, safety expectations, and speed expectations.

[0020] Furthermore, in step S3, the regression analysis includes the following:

[0021] Categorical variables include, but are not limited to, gender, visual acuity, driving age, number of days a week spent riding an electric bicycle, visual load, behavioral load, safety expectations, and speed expectations. In regression modeling, categorical variables should be distinguished from numerical variables; therefore, they are converted into variables taking values ​​of 0 or 1, i.e., dummy variables. Multiple dummy variables exhibit a completely linear relationship; therefore, if a categorical variable has k class values, only the first k-1 dummy variables need to be introduced into the model. Dummy variable groups cannot be modeled independently; therefore, multiple regression models are performed using an input method, and the optimal regression model is obtained by selecting the variables.

[0022] Furthermore, in step S4, establishing a linear regression model for the reaction time of an electric bicycle includes the following:

[0023] Speed ​​and dummy variables include, but are not limited to, whether a full-face helmet is worn, whether a 3 / 4 helmet is worn, whether an electric bicycle is ridden every day of the week, whether an electric bicycle is ridden 2-1 days of the week, whether safety expectations are very high, and whether safety expectations are low.

[0024] A linear regression model for the reaction time of an electric bicycle is established based on speed and dummy variables. The specific formula is as follows:

[0025] y=449.034+21.344X1+214.532X2+76.170X3-125.641X4-91.383X5-184.556X6+132.493X7

[0026] Where y represents the electric bicycle's reaction time, X1 represents speed, X2 represents whether a full-face helmet is worn, X3 represents whether a 3 / 4 helmet is worn, X4 represents whether the electric bicycle is ridden every day of the week, X5 represents whether the electric bicycle is ridden 2-1 days of the week, X6 represents whether the safety expectation is very high, and X7 represents whether the safety expectation is very low; the unit of reaction time is milliseconds.

[0027] Furthermore, the goodness-of-fit test, significance test of the regression equation, significance test of the regression coefficient, and residual analysis were performed on the linear regression model of the electric bicycle reaction time. The test results show that the model is reliable.

[0028] Furthermore, the present invention also proposes an electric bicycle rider reaction time prediction system, including...

[0029] The data acquisition module is used to design test schemes for electric bicycle reaction time, conduct real-vehicle tests on closed road sections and record videos, and collect data on electric bicycle riders wearing helmets, riding speed, riding steering and reaction time based on the video recordings.

[0030] The variable collection module is used to design questionnaires to collect variables that affect the reaction time of cyclists participating in the experiment when riding electric bicycles.

[0031] The linear regression model building module is used to screen influencing variables and convert categorical variables into dummy variables, and to build a linear regression model of electric bicycle reaction time with speed and dummy variables as independent variables.

[0032] The reaction time prediction module is used to input the acquired variables into the linear regression model of electric bicycle reaction time to predict the reaction time.

[0033] Furthermore, the present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the electric bicycle rider reaction time prediction method described above.

[0034] The present invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, performs the electric bicycle rider reaction time prediction method described above.

[0035] The present invention adopts the above technical solution, and its significant technical effects compared with the prior art are as follows:

[0036] This invention innovatively proposes the concept of electric bicycle reaction time, comprehensively considering the influence of factors such as the rider's helmet, riding speed, steering, physiological characteristics, behavioral load, and driving expectations on the rider's reaction time. It further constructs a rider reaction time prediction model, providing a theoretical basis for calculating the line of sight and collision distance of electric bicycles, which is of great significance for ensuring the safety of electric bicycles. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the overall implementation of the present invention.

[0038] Figure 2 This is a schematic diagram of the experimental method in an embodiment of the present invention.

[0039] Figure 3 This is a diagram showing the difference in reaction time between different helmets when turning left in an embodiment of the present invention.

[0040] Figure 4 This is a diagram showing the difference in reaction time between different helmets when turning right in an embodiment of the present invention.

[0041] Figure 5 This is a graph showing the difference in reaction time between different genders in an embodiment of the present invention.

[0042] Figure 6 This is a graph showing the difference in reaction time under different visual conditions in an embodiment of the present invention.

[0043] Figure 7 This is a graph showing the difference in reaction time for different driving days per week in an embodiment of the present invention.

[0044] Figure 8 This is a graph showing the difference in reaction time among drivers with different driving experience in an embodiment of the present invention.

[0045] Figure 9 This is a graph showing the difference in reaction time under different visual loads in an embodiment of the present invention.

[0046] Figure 10 This is a graph showing the difference in reaction time for different safety expectations in embodiments of the present invention.

[0047] Figure 11 This is a graph showing the relationship between speed and reaction time in an embodiment of the present invention.

[0048] Figure 12 This is a residual analysis diagram in an embodiment of the present invention. Detailed Implementation

[0049] To illustrate the technical content, structural features, objectives, and effects of this invention in detail, the following description, in conjunction with accompanying drawings and examples, will provide a detailed explanation of the invention.

[0050] To achieve the above objectives, this invention proposes a method for predicting the reaction time of electric bicycle riders, such as... Figure 1 As shown, it includes the following steps:

[0051] S1. Design a test plan for the reaction time of electric bicycles, conduct real-vehicle tests on closed roads and record them. Based on the video recordings, collect data on the electric bicycle rider's helmet wearing, riding speed, steering, and reaction time. Specific details include:

[0052] In this embodiment, 28 university students aged 20-28 (average 23.5) with varying electric bicycle riding experience were recruited. Due to the different experimental scenarios, each participant underwent two sets of experiments: one set involving left turns and one set involving right turns. When turning left, participants wore different helmets, including full-face helmets, 3 / 4 half-helmets, 1 / 2 half-helmets, and no helmet. The experiments were conducted on closed sections of the campus to ensure no interference from other vehicles or pedestrians. Experiments were only conducted during periods of clear weather. Before the experiments, participants were asked relevant questions to confirm their good physical and mental condition, eliminating interference from road conditions, weather, and psychological factors. Specific methods are as follows... Figure 2As shown, test subjects first rode on sections of road where obstacles were not visible, wearing different helmets, and then prepared to turn left or right at intersections with poor visibility. Due to building obstruction, the blue triangular cone obstacle was only observed by the test subjects when they reached a certain position near the intersection; this point was called the obstacle visibility point. During the experiment, the time from when the test subject rode to the obstacle visibility point to when they moved their hand to the brake was the electric bicycle's reaction time. During the experiment, the position of the obstacle would move back and forth, and the position of the obstacle visibility point would also change accordingly to ensure the reliability of the data. The testing procedure was as follows: ① Before the start of the experiment, the test subjects filled out a questionnaire to investigate their physiological attributes, driving experience, etc., and read the experimental instructions. ② The staff introduced the test vehicle and informed them of relevant precautions. The test subjects conducted a pre-drive to familiarize themselves with the test vehicle and the test road to ensure the accuracy of the data. ③ The staff informed the test subjects of the specific experimental procedure and relevant precautions, and told them to maintain normal driving habits. There was no speed limit during the experiment, but the experiment would be terminated in case of an emergency. To better obtain data, the experiment stipulated that the rider's fingers should not be placed on the brake before the obstacle was detected. Therefore, the reaction time tested in this paper includes the time it takes for the finger to move to the brake. ④ Staff randomly moved obstacles and marked their visible points. The method for marking these visible points was not explained to the test subjects to ensure data accuracy. At the start of the experiment, the test subjects followed the procedure and completed a driving experience questionnaire afterward.

[0053] After the experiment, the recorded video was processed to capture the reaction time of the electric bicycles. Reaction time refers to the time from when the subject moves to the point of visibility of the obstacle to when the subject places their hand on the brake and begins to press it. After obtaining the reaction time, outliers were removed using the Puata criterion, and a normality test was performed. The braking reaction times of all groups of electric bicycles conformed to a normal distribution.

[0054] S2. Through relevant literature review, factors that may affect the reaction time of electric bicycles include rider's physiological attributes, riding experience, riding load, and riding expectations; design a questionnaire to collect data on the physiological characteristics, riding load, and riding expectations of the riders participating in the experiment.

[0055] The received questionnaires underwent reliability and validity testing, and the results were satisfactory. Physiological attributes included age, gender, and vision; cycling experience included the number of days a week the rider had ridden an electric bicycle; cycling load included visual load, i.e., whether the rider believed that the helmet would affect their vision while riding; cycling expectations referred to safety expectations, which are the degree of importance the rider places on cycling safety during the ride, with higher safety expectations indicating a greater emphasis on cycling safety.

[0056] S3. Based on steps S1 and S2, further screen influencing variables and convert categorical variables into dummy variables for regression analysis. The specific steps are as follows:

[0057] Since the test individuals remained unchanged, to further analyze the impact of different helmets on driver reaction time, analysis of variance (ANOVA) was used to test the significance of the data. A combination of box plots, line plots, and ANOVA significance testing was used for analysis. The results showed that different helmets do affect driver reaction time. The results of multiple comparisons are represented using letter notation, such as... Figure 3 As shown, there were no significant differences in landscape response times when the same letter was used, but significant differences were observed when different letters were used.

[0058] By setting up left and right turns, and comparing different helmets pairwise, the study explored the significant impact of different helmets on the reaction time of electric bicycle riders. Figure 3 As shown, in the left-turn situation, there were significant differences in reaction time among different helmet types. The reaction time of e-bike riders wearing full-face helmets was significantly longer than that of riders wearing 3 / 4 helmets (p<0.05), half-helmets (p<0.01), and those without helmets (p<0.01). There were no significant differences in reaction time among e-bike riders wearing 3 / 4 helmets, half-helmets, and full-face helmets. Figure 4 As shown in Table 1, during right turns, reaction times differed significantly for most helmets. The reaction time for e-bike riders wearing full-face helmets was significantly longer than that for those wearing 3 / 4 helmets (p<0.05), half helmets (p<0.01), and no helmets (partial helmets, p<0.01). The reaction time for e-bike riders wearing 3 / 4 helmets was significantly longer than that for those without helmets (p<0.05). There was no significant difference in reaction time between e-bike riders wearing half helmets and those without helmets.

[0059] Table 1. Reaction times of different helmets on left and right turns.

[0060] Full-face helmet 960.6059 973.6831 3 / 4 helmet 734.1670 773.1478 Half helmet 672.5005 647.2217 none 611.4037 612.2805

[0061] From Table 1, Figure 3 and Figure 4 It can be seen that wearing different helmets affects the reaction time of e-bike riders. Full-face helmets result in the longest reaction time, followed by 3 / 4 helmets and half-helmets, with no helmet being the least effective. This is likely because full-face helmets obstruct the rider's field of vision. Although regulations specify helmet field of vision, full-face helmets still have some impact on reaction time. However, it should be noted that while there are significant differences in reaction time in some situations, the numerical differences are not substantial.

[0062] Physiological attributes included age, sex, and vision. Since the participants were mostly between 20 and 30 years old, the differences were small and therefore not analyzed. Women comprised 43.5% of the participants, and men comprised 56.5%. Vision levels were categorized into three levels: average, good, and very good. Average vision accounted for 9.8%, good vision for 75.5%, and very good vision for 14.7%. Descriptive statistics are shown in Table 2, indicating that men's reaction times were significantly shorter than women's. The average reaction time for those with very good vision was shorter than that for those with good or average vision.

[0063] Table 2 Reaction Time of Electric Bicycles with Different Physiological Attributes

[0064]

[0065] Analysis of variance (ANOVA) was used to test the differences in different physiological attributes. Although the variances among groups were unequal, the Welch method could be used to compare the results. Overall, there were significant differences between genders (p<0.01) and between different visual acuity conditions (p<0.01). See below for detailed pairwise comparison results. Figure 5 and Figure 6 .

[0066] from Figure 5 It can be seen that gender has a significant impact on reaction time, with women's reaction times being significantly longer than men's. This may be because men have a better sense of control and are more composed in the face of unexpected situations. Figure 6 As shown, the reaction time of the group with good vision was significantly longer than that of the group with average vision. The survey suggests that vision status is a subjective assessment by the test takers, and those who chose average vision were more proactive and flexible, thus resulting in shorter reaction times.

[0067] Regarding driving habits such as driving experience, driving load, and expectations, statistical results showed that riders who rarely ride e-bikes each week had the longest reaction times. As driving experience increased, riders' reaction times first increased and then decreased, with the longest reaction times observed in riders with 2-5 years of experience, and the shortest reaction times observed in riders with less than 1 year or more than 10 years of experience. Regarding driving load, the group that considered helmets to have a significant impact on vision and driving behavior had the shortest mean reaction times. Regarding safety expectations, higher safety expectations resulted in shorter reaction times. Analysis of variance was used to further investigate the differences among groups with different driving habits, yielding multiple comparison results as follows: Figure 7 , Figure 8 , Figure 9 and Figure 10 As shown.

[0068] like Figure 7As shown, there are significant differences in reaction time among groups with different driving frequencies. The group with a weekly driving frequency of almost zero has a significantly shorter reaction time than others, possibly because their driving frequency is relatively low. Visual load, behavioral load, speed expectations, and safety expectations also have a significant impact on reaction time. Figure 8 As shown, there are significant differences in reaction time among electric bicycle riders with different experience levels. The reaction time of riders with less than one year of experience or more than ten years of experience is significantly shorter than that of other groups. This may be because beginners are more alert and have shorter reaction times, while experienced riders have more time and therefore shorter reaction times. Figure 9 As shown, regarding visual load, cyclists who believed that helmets did not significantly affect their driving vision had significantly shorter reaction times than other groups. Figure 10 As shown, there were significant differences between groups in terms of safety expectations. The reaction time of the groups with average and low safety expectations was significantly longer than that of the groups with high and very high safety expectations.

[0069] Speed ​​refers to the speed at which the rider is about to see an obstacle during the test and before braking. For example... Figure 11 As shown, a rough estimate indicates a strong positive correlation between speed and reaction time. To further analyze the relationship between speed and reaction time, the Pearson correlation coefficient was used to measure their statistical relationship. Using SPSS software, the correlation coefficient between speed and reaction time was 0.547, with a significance probability less than 0.01. Therefore, it can be concluded that speed and reaction time are positively correlated.

[0070] In this embodiment, most independent variables are categorical variables. When modeling regression, these should be distinguished from numerical independent variables. Therefore, the categorical explanatory variables are transformed into several variables with values ​​of 0 or 1; these are called dummy variables. A perfect linear relationship exists between multiple dummy variables. Therefore, if a categorical independent variable has k class values, only the first k-1 dummy variables need to be introduced into the model. Dummy variable groups cannot be modeled independently; therefore, the input method is used to model multiple times to obtain the optimal model.

[0071] SPSS software was used for testing, and finally, helmet type, safety expectation, number of driving days per week and speed were incorporated into the model from the categorical variables.

[0072] S4. Establish a linear regression model for electric bicycle reaction time using speed and dummy variables (including: whether a full-face helmet is worn, whether a 3 / 4 helmet is worn, whether the electric bicycle is ridden every day of the week, whether the electric bicycle is ridden 2-1 days of the week, whether the safety expectation is very high, and whether the safety expectation is low) as independent variables. The specific formula is as follows:

[0073]

[0074] The regression analysis results are shown in Tables 3 and 4.

[0075] Table 3. Results of Multiple Linear Regression

[0076] 0.846 0.716 0.693 113.304 31.60 <0.001 1.821

[0077] After establishing the regression equation, various statistical tests are needed to select the optimal model, mainly including goodness-of-fit tests, regression equation significance tests, regression coefficient significance tests, and residual analysis. The goodness-of-fit test examines the density of sample data points clustered around the regression line; in multivariate regression analysis, adjusted R² is typically used. 2 As an indicator of model fit, the adjusted R-squared value is... 2 The larger the value, the higher the model fit. As shown in Table 3, a high model fit (adjusted R value) indicates a good model fit. 2 =0.693). The significance test examines whether the linear relationship between the explained variable and the explanatory variables is significant, and the F-statistic is generally used for the test.

[0078] Table 4. Parameters of the Multiple Regression Model

[0079]

[0080]

[0081] Table 4 shows that the regression equation of the model is significant (F = 31.60, P < 0.001). The significance test of the regression coefficients studies whether each explanatory variable can effectively explain the linear change of the explained variable. Generally, the T-statistic is used. Table 4 shows that the regression coefficient of the continuous variable speed is significant (P < 0.01). Among the categorical variables, the regression coefficients of six variables—whether to wear a full-face helmet, whether to wear a 3 / 4 helmet, whether to ride an electric bicycle every day of the week, whether to ride an electric bicycle 2-1 days of the week, whether the safety expectation is very high, and whether the safety expectation is low—are significant (P < 0.01).

[0082] Residual analysis includes normality analysis, independence analysis, and heteroscedasticity analysis. A normality test was performed on the residuals, and no significant difference was found between the residuals and a normal distribution. Further analysis was conducted to determine multicollinearity among the independent variables. Table 4 shows that the DW values ​​are close to 2, indicating weak correlation among the residuals. The tolerances of the independent variables are all greater than 0.1, and the variance inflation factor (VIF) is less than 5, indicating that there is no multicollinearity between the independent and dependent variables. The standardized predicted values ​​of the regression equation and the scatter plot of the standardized residuals are observed. Figure 12 As shown, there is no heteroscedasticity. Based on the above analysis, the model is reliable.

[0083] This invention also proposes an electric bicycle rider reaction time prediction system, including a data acquisition module, a variable acquisition module, a linear regression model construction module, a reaction time prediction module, and a computer program that can run on a processor. It should be noted that each module in the above system corresponds to a specific step of the method provided in this invention embodiment, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention embodiment.

[0084] This invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. It should be noted that when the processor executes the computer program, it corresponds to the specific steps of the method provided in this invention, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention.

[0085] This invention also proposes a computer-readable storage medium storing a computer program. It should be noted that when the computer program is executed by a processor, it corresponds to the specific steps of the method provided in this invention, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention.

[0086] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method of predicting the reaction time of an e-bike rider, characterized in that, include: S1. Design a test plan for the reaction time of electric bicycles, conduct real vehicle tests and record videos on closed road sections, and collect data on the electric bicycle rider's helmet wearing, riding speed, riding steering and reaction time based on the video recordings; S2. Design a questionnaire to collect variables that affect the reaction time of cyclists participating in the experiment when riding electric bicycles; S3. Based on steps S1 and S2, further screen the influencing variables and convert the categorical variables into dummy variables for regression analysis; Among them, the categorical variables include gender, visual acuity, driving age, number of days in a week of riding an electric bicycle, visual load, behavioral load, safety expectation, and speed expectation; Speed ​​and dummy variables include whether a full-face helmet is worn, whether a 3 / 4 helmet is worn, whether the e-bike is ridden every day of the week, whether the e-bike is ridden 1-2 days of the week, whether the safety expectation is very high, and whether the safety expectation is low. S4. Establish a linear regression model of electric bicycle reaction time using speed and dummy variables as independent variables; S5. Input the obtained variables into the linear regression model of electric bicycle reaction time to predict the reaction time.

2. The electric bicycle rider reaction time prediction method according to claim 1, characterized in that, In step S1, the collected data includes the following: The rider rides under different types of helmets and turning conditions. When an obstacle is detected, the rider brakes to slow down until coming to a stop, and this process is recorded by a video recording device. By playing back the video recording, the time from when the obstacle is detected to when the brake is applied is recorded as the rider's reaction time. The distance before the obstacle is detected is measured, and the time taken to travel that distance is read from the video to calculate the riding speed. Among them, the electric bicycle rider's reaction time is the selected reaction time, which includes the time it takes for the rider's finger to move to the brake.

3. The electric bicycle rider reaction time prediction method of claim 2, wherein, In step S1, the helmets worn by the cyclist include full-face helmets, 3 / 4 half helmets, 1 / 2 half helmets, and no helmet; cycling turns include left turns and right turns.

4. The electric bicycle rider reaction time prediction method of claim 3, wherein, In step S2, the variables collected include the following: Variables affecting the reaction time of electric bicycles include physiological attributes, riding experience, riding load, and riding expectations; Physiological attributes include age, gender, and vision; cycling experience includes the number of days a week spent riding an e-bike; cycling load includes visual load; and cycling expectations include safety expectations.

5. The electric bicycle rider reaction time prediction method of claim 4, wherein, Step S3 involves regression analysis, including the following: Categorical variables are transformed into variables that take values ​​of 0 or 1, i.e., dummy variables; multiple regression models are then performed using the input method.

6. The electric bicycle rider reaction time prediction method of claim 5, wherein, In step S4, establishing a linear regression model for the reaction time of an electric bicycle includes the following: A linear regression model for the reaction time of an electric bicycle is established based on speed and dummy variables. The specific formula is as follows: ； wherein, indicates the reaction time of the e-bike, indicates the speed, indicates whether a full helmet is worn, indicates whether a 3 / 4 helmet is worn, indicates whether the e-bike is driven every day of the week, indicates whether the e-bike is driven 1-2 days of the week, indicates whether the safety expectation is very high, indicates whether the safety expectation is lower.

7. A system applied to the electric bicycle rider reaction time prediction method of claim 1, characterized in that, include The data acquisition module is used to design test schemes for electric bicycle reaction time, conduct real-vehicle tests on closed road sections and record videos, and collect data on electric bicycle riders wearing helmets, riding speed, riding steering and reaction time based on the video recordings; The variable collection module is used to design questionnaires to collect variables that affect the reaction time of cyclists participating in the experiment when riding electric bicycles; The linear regression model building module is used to screen influencing variables and convert categorical variables into dummy variables, and to build a linear regression model of electric bicycle reaction time with speed and dummy variables as independent variables; The reaction time prediction module is used to input the acquired variables into the linear regression model of electric bicycle reaction time to predict the reaction time.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to perform the method of any one of claims 1 to 6.

Images