Method for determining speed limit value of small-radius circular curve of mountainous highway

Abstract

The mountainous area highway small-radius circular curve speed limit value determination method belongs to the technical field of small-radius circular curve speed limit value determination, and solves the technical problem that the prior art only determines the speed limit value according to the vehicle dynamics parameters, without considering the physiological load characteristics of the driver in the mountainous area small-radius curve, resulting in insufficient adaptability of the speed limit result. By introducing a plurality of physiological indexes including heart rate, skin electricity and muscle electricity to construct a comprehensive physiological load index model, the comprehensive physiological load index is calculated by using the comprehensive physiological load index model, the moderate load interval of the comprehensive physiological load index is taken as a speed limit value screening criterion, and the recommended speed limit value is determined. The present application is used for determining the speed limit value of the mountainous area highway small-radius curve, and improves the driving safety and efficiency.

Description

Technical Field

[0001] This invention relates to the technical field of determining speed limits for small-radius circular curves, specifically to a method for determining speed limits for small-radius circular curves on mountain roads. Background Technology

[0002] Highways play a vital role in the national economy, serving not only as a core component of the comprehensive transportation system but also directly impacting economic development and the convenience of people's travel.

[0003] The construction of mountain roads is often constrained by natural conditions, resulting in lower design standards and some road alignment parameters approaching their limits. This makes driving difficult and significantly increases the accident rate, especially in seasonally frozen areas such as Northeast and Northwest China. Freeze-thaw cycles cause periodic deterioration of the road surface structure, and winter snow and ice cover further reduces the road surface friction coefficient, exacerbating safety hazards on mountain roads. Current speed limits are determined based on road geometry and vehicle dynamics principles, and a uniform speed limit is usually applied to the same road. This is insufficient to accommodate the varying horizontal and vertical parameters of different road sections in mountainous areas due to terrain constraints. The safety risks are particularly prominent on small-radius circular curve sections due to their lower speed limits.

[0004] Existing technologies are primarily based on vehicle dynamics principles and are suitable for relatively simple traffic environments such as ramps. However, they do not consider the complex human factors unique to mountainous roads, such as limited visibility, roadside interference, and driver psychological and physiological stress. On small-radius horizontal curves in mountainous areas, the combined effects of insufficient visibility and operational complexity lead to driver psychological stress levels far exceeding what physical parameters can predict. Directly applying these technologies to determine speed limits on mountainous roads would result in speed limits that, while meeting mechanical safety requirements, exceed the driver's physiological tolerance threshold, demonstrating insufficient adaptability.

[0005] In summary, existing technologies suffer from the problem of insufficient adaptability in determining speed limits based solely on vehicle dynamics parameters without considering the physiological load characteristics of drivers on small-radius curves in mountainous areas. Summary of the Invention

[0006] This invention solves the technical problem that existing technologies determine speed limits based solely on vehicle dynamics parameters without considering the physiological load characteristics of drivers on small-radius curves in mountainous areas, resulting in insufficient adaptability of speed limit results.

[0007] The method for constructing a comprehensive physiological load index model for determining speed limits, as described in this invention, includes the following steps: Step 1: Acquire multiple physiological indicators of multiple drivers as they pass through the target road section by using a multi-channel physiological signal acquisition device worn on the driver's body; Step 2: Process the collected multiple physiological indicators, calculate the reliability and individual variability of different physiological indicators based on the processed data, and statistically analyze the safety thresholds corresponding to different physiological indicators. Step 3: Based on the safety thresholds, reliability, and individual variability corresponding to different physiological indicators, construct a comprehensive physiological load index model.

[0008] Furthermore, in one embodiment of the present invention, the comprehensive physiological load index model specifically comprises: ; in, For heart rate growth rate, For skin cell growth rate, This is the root mean square value of electromyography. This is a comprehensive physiological load index.

[0009] The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to the present invention is implemented by the aforementioned comprehensive physiological load index model construction method for determining the speed limit value, and includes the following steps: Step A1: For different radii of circular curves on mountain roads, preset multiple vehicle speeds respectively; Step A2: Obtain multiple physiological indicators of the driver when passing through different circular curve radii on mountain roads at different vehicle speeds; Step A3: Use the multiple physiological indicators obtained in step A2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value for the radius of the corresponding circular curve of the mountain road based on the comprehensive physiological load index.

[0010] Furthermore, in one embodiment of the present invention, in step A1, multiple vehicle operating speeds are preset for different radii of circular curves on mountain roads, specifically as follows: When the radius of the circular curve on the mountain road is 60m, the preset vehicle speeds are 30km / h, 35km / h and 40km / h. When the radius of the circular curve on the mountain road is 70m, the preset vehicle speeds are 35km / h, 40km / h and 45km / h. When the radius of the circular curve on the mountain road is 80m, the preset vehicle speeds are 40km / h, 45km / h and 50km / h. When the radius of the circular curve on the mountain road is 90m and 100m, the preset vehicle speed is 45km / h, 50km / h and 55km / h.

[0011] Furthermore, in one embodiment of the present invention, in step A3, determining the recommended speed limit value for the radius of the corresponding circular curve of a mountain road based on the comprehensive physiological load index specifically involves: The comprehensive physiological load index is divided into multiple interval levels, and the vehicle operating speed that keeps the comprehensive physiological load index within the moderate load range is selected as the recommended speed limit value for the radius of the corresponding circular curve of the mountain road.

[0012] Furthermore, in one embodiment of the present invention, the division of the comprehensive physiological load index into multiple interval levels specifically includes: when When it is in the light load range, when When the time is within the moderate load range, During the high-load period, when This is the overload zone.

[0013] Furthermore, in one embodiment of the present invention, when the radius of the circular curve of the mountain road is 60 meters, the recommended speed limit is 35 km / h; when the radius of the circular curve of the mountain road is 70 meters, the recommended speed limit is 40 km / h; when the radius of the circular curve of the mountain road is 80 meters, the recommended speed limit is 45 km / h; when the radius of the circular curve of the mountain road is 90 meters, the recommended speed limit is 50 km / h; and when the radius of the circular curve of the mountain road is 100 meters, the recommended speed limit is 55 km / h.

[0014] The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to the present invention is based on the above-mentioned comprehensive physiological load index model construction method for determining the speed limit value, and includes the following steps: Step B1: Obtain the radius of the circular curve of the mountain road and preset different vehicle speeds; Step B2: Construct a physiological index prediction model. Based on the radius of the circular curve of the mountain road and the preset different vehicle operating speeds, calculate the physiological parameters corresponding to the vehicle operating speed. Step B3: Use the physiological parameters calculated in step B2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value corresponding to the radius of the circular curve of the mountain road based on the comprehensive physiological load index.

[0015] Furthermore, in one embodiment of the present invention, the physiological indicator prediction model in step B2 includes a heart rate growth rate prediction model, a skin conductance growth rate prediction model, and an electromyography root mean square value prediction model.

[0016] Furthermore, in one embodiment of the present invention, the heart rate growth rate prediction model specifically comprises: ; The skin cell growth rate prediction model is as follows: ; The root mean square value prediction model for electromyography is as follows: ; in, The radius of the circular curve is... The vehicle's operating speed.

[0017] This invention solves the technical problem of existing technologies that determine speed limits solely based on vehicle dynamics parameters, neglecting the physiological load characteristics of drivers on small-radius curves in mountainous areas, resulting in insufficient adaptability of the speed limits. Specific beneficial effects of this invention include: 1. This invention proposes a comprehensive physiological load index model construction method for determining speed limits. By introducing multiple physiological indicators, including heart rate, skin conductance, and electromyography, a comprehensive physiological load index model is constructed. This overcomes the one-sidedness of a single physiological indicator and can more comprehensively assess the driver's physiological load level during driving on small-radius circular curves. It extends the determination of speed limits from the traditional vehicle physical motion state level to the driver's physiological tolerance level, solving the technical problem of insufficient adaptability of existing pure physical dynamic speed limit methods on small-radius curved road sections, and making the recommended speed limit more consistent with the actual physiological load characteristics of the driver. 2. This invention proposes a method for determining the speed limit value of small-radius circular curves on mountain roads, such as... Figure 1 As shown, based on driving simulation, multiple physiological indicators such as heart rate, skin conductance, and electromyography are directly collected. The moderate load range of the comprehensive physiological load index is used as the screening criterion for speed limit value. This can maximize road traffic efficiency while ensuring the physiological safety of drivers and provide a unified quantitative judgment standard for differentiated speed limits of circular curves with different radii. 3. This invention proposes a method for determining the speed limit value of small-radius circular curves on mountain roads, such as... Figure 1 As shown, a physiological index prediction model can be constructed, which can quickly calculate the speed limit value by only inputting the radius and speed, greatly reducing the test cost and making it suitable for rapid evaluation of large-scale road networks. 4. This invention proposes a method for determining speed limits for small-radius circular curves on mountain roads. It provides specific recommended speed limits for five standard radius circular curves on mountain roads and offers a linear interpolation method for determining the transition radius. This method can be directly applied to road speed limit sign settings or variable speed limit control systems, and has the characteristics of strong engineering practicality and easy promotion. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of the method for determining the speed limit value of a small radius circular curve on a mountain highway as described in this invention; Figure 2 This is a schematic diagram illustrating the distribution of driver physiological indicators (taking heart rate as an example) as described in Implementation Method 1; wherein, Figure 2 (a) Heart rate distribution of 5 drivers under 15 test conditions; Figure 2 (b) Data distribution for each driver after standardization using the growth rate; Figure 2 (c) All 75 sets of data show a right-skewed distribution; Figure 3 This is a schematic diagram of the inter-individual coefficient of variation of physiological indicators under different experimental conditions as described in Implementation Method 1; Figure 4 This is a schematic diagram of the PLI distribution characteristics of circular curves with different radii at three test speeds as described in Implementation Method 2. Detailed Implementation

[0019] Various embodiments of the present invention will now be clearly and completely described with reference to the accompanying drawings. The embodiments described with reference to the drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0020] Implementation Method 1: A method for constructing a comprehensive physiological load index model for determining speed limit values, comprising the following steps: Step 1: Acquire multiple physiological indicators of multiple drivers as they pass through the target road section by using a multi-channel physiological signal acquisition device worn on the driver's body; Step 2: After processing the collected multiple physiological indicators, calculate the reliability and individual variability of different physiological indicators, and statistically analyze the safety thresholds corresponding to different physiological indicators. Step 3: Based on the safety thresholds, reliability, and individual variability corresponding to different physiological indicators, construct a comprehensive physiological load index model.

[0021] In road traffic systems, the behavior and state of traffic participants play a decisive role in driving safety. Mountain roads with small-radius circular curves have complex geometry, requiring drivers to complete a series of actions—including curve judgment, speed adjustment, and steering—within a short time, resulting in a much higher operational load than on straight sections. Insufficient estimation of curve curvature or excessive entry speed can easily cause vehicles to deviate from their intended trajectory, even skidding or running off the road. Furthermore, the driving environment on small-radius circular curves on mountain roads significantly impacts drivers' physiological indicators, often subjecting them to greater mental stress and tension. The smaller the radius of the horizontal curve, the higher the complexity of driving operations. Excessive vehicle speed further exacerbates driver tension, increasing the risk of traffic accidents. Therefore, in-depth analysis of drivers' operational characteristics and physiological response patterns on small-radius horizontal curves is crucial for establishing reasonable speed limits.

[0022] Current highway speed limits are primarily determined based on road geometry and vehicle dynamics principles. This method, grounded in road alignment characteristics and vehicle performance, provides a unified basis for speed limit setting. However, it has shortcomings in practical applications on mountainous highways. On the one hand, the traditional method treats drivers as idealized subjects without individual differences, failing to consider their physiological and psychological reactions to complex driving environments. On the other hand, while some highway sections use uniform speed limits, the radii of curves along the route vary greatly. Lower speed limits on large-radius sections affect traffic efficiency, while applying the same speed limit on small-radius sections poses safety hazards. This situation is particularly pronounced on Class III highways. As an important component of rural transportation networks, Class III highways in mountainous areas, while ensuring regional connectivity, also face serious safety challenges due to their lower technical standards.

[0023] To address the aforementioned technical problems, this embodiment proposes a method for constructing a comprehensive physiological load index model for determining speed limits, specifically including the following steps: Step 1: Acquire multiple physiological indicators of multiple drivers as they pass through the target road section by using a multi-channel physiological signal acquisition device worn on the driver's body; This implementation method uses the VTD (Virtual Test Drive) platform to build a test scenario to simulate the driving environment of small-radius circular curve sections on mountain roads. This implementation method designs five circular curve scenarios with different radii: 60 meters, 70 meters, 80 meters, 90 meters, and 100 meters.

[0024] The driving simulator used in this embodiment comprises two main components: hardware and software. The hardware system includes a cockpit, a six-degree-of-freedom dynamic platform, a visual display device, sound simulation equipment, an information acquisition module, and a control and computing system. The cockpit is modified from a real car, fully preserving the in-vehicle control environment and creating a realistic driving experience for the driver. The dynamic platform houses the entire cockpit and provides multi-dimensional motion simulation, accurately reproducing the vehicle's movement in curves. The visual display device is equipped with three sets of projection equipment and a display terminal. The projection equipment works together to construct a 180° field of view to display the road conditions ahead and to the sides, while the display terminal is located at the rearview mirror to present the rear view. The information acquisition module records data through sensors placed in locations such as the pedals, gear lever, and steering wheel. The sound simulation equipment can reproduce various driving-related sound effects such as engine sounds and ambient noise.

[0025] The simulator software system uses the VTD (Virtual Test Drive) platform. VTD platform simulation requires two stages: static environment construction and dynamic scenario setting. Static environment construction is achieved using the ROD (Road Designer) tool, while dynamic scenario setting is completed using the SE (Scenario Editor). This implementation uses this system to construct five mountain road curve scenarios with different radii (60 meters, 70 meters, 80 meters, 90 meters, and 100 meters) for subsequent collection of driver physiological data.

[0026] The physiological data acquisition equipment used was the BIOPAC physiological instrument, and the acquired data was processed using AcqKnowledge 5.0 (physiological signal acquisition and analysis software).

[0027] Data collection plan development: 1. Theoretical basis of velocity combination scheme For small-radius circular curves within the radius range of 60~100m, the selection of test speed combinations directly affects the accuracy of identifying threshold points for physiological indicator changes, and also relates to the reliability of the speed limit derivation results.

[0028] (1) Design concept of speed selection The speed gradient is set at 5 km / h. This ensures that the difference between adjacent speed levels can be clearly perceived by the driver, and also provides sufficient data support for the subsequent derivation of speed limits.

[0029] (2) Lateral acceleration classification and vehicle dynamics basis Lateral acceleration is the core physical quantity for assessing driving load in this implementation method. The greater the lateral acceleration, the more pronounced the driver's discomfort symptoms, and the greater the changes in physiological parameters such as heart rate and skin conductance, reflecting significant activation of the sympathetic nervous system. During cornering, drivers counteract vehicle sway to maintain head stability through muscle contraction; the degree of muscle activity is directly related to the magnitude of lateral acceleration. Furthermore, drivers tend to actively reduce speed in high-speed corners to decrease lateral acceleration; this behavioral characteristic can be described using a safety margin control model. Lateral acceleration can also serve as an indicator for assessing the hazard level of a driving behavior, providing a physical reference for subsequently constructing physiological load indicators using physiological parameters such as electrocardiogram, skin conductance, and electromyography.

[0030] According to the formula for circular motion Different combinations of velocity and radius correspond to different levels of lateral acceleration, among which, It is lateral acceleration. For vehicle operating speed, The radius of the circular curve is denoted as . This implementation requires identifying the critical threshold of the driver's physiological response; therefore, the speed-radius combination must be designed to cover the entire load range from comfortable to uncomfortable. The lateral acceleration produced by each combination ranges from 1.16 to 2.60 m / s². 2 The system covers driving conditions with different comfort levels. By testing the physiological response limits of drivers under high lateral acceleration conditions, it can identify the critical points of individual differences, obtain the physiological response thresholds of drivers, and determine the speed limits for mountain roads based on these thresholds.

[0031] (3) Speed ​​setting based on operating speed theory This implementation method sets up a test range of 30 to 55 km / h, which can cover different driving styles and road condition adaptability, providing a sufficient experimental basis for establishing a speed limit scheme based on physiological load.

[0032] 2. Differentiated speed combination strategy Currently, mountain roads mostly use uniform speed limits, applying the same limit regardless of the radius of the circular curve, failing to differentiate the actual risk differences between road sections with different radii. Based on the safety characteristics and risk levels of circular curves with different radii, this implementation method adopts a differentiated speed testing strategy, enabling the collected data to reflect the true changes in the driver's physiological load under conditions of each radius.

[0033] (1) 60m radius velocity test plan The minimum radius is 60m, and the tests focus on three speed levels: 30, 35, and 40 km / h, to verify the safety baseline under extreme conditions. In the three speed tests within the 60-meter radius, the lateral acceleration ranges from 1.16 to 2.06 m / s², covering the comfort zone to the critical transition zone. The test data at these three speed levels can reflect the physiological load changes of the driver under these road conditions.

[0034] (2) 70m and 80m radius speed test plan The 70m radius test falls between the extreme and general conditions, with test speeds of 35, 40, and 45 km / h. For the 80m radius test, the speed range is shifted upwards to 40, 45, and 50 km / h. The lateral acceleration at the 70m radius at the three test speeds of 35–45 km / h is 1.36–2.23 m / s², and at the 80m radius at 40–50 km / h it is 1.54–2.41 m / s². Both sets of tests cover both the comfortable driving range and the critical range, facilitating the observation of physiological responses within the transition radius.

[0035] (3) 90-meter and 100-meter radius speed test plan For general radii of 90m and 100m, tests were conducted at high speeds of 45, 50, and 55 km / h to verify speed adaptability under near-standard design conditions. The lateral accelerations at speeds of 45, 50, and 55 km / h for radii of 90m and 100m were 1.74–2.59 m / s² and 1.74–2.33 m / s², respectively, which adequately tested the driver's physiological adaptability at higher speeds.

[0036] (4) Combinatorial optimization The speed-radius combination scheme in this implementation does not test all five speed levels for all radii. Instead, it selects three test speeds specifically based on the risk characteristics of each radius, simplifying the 25 combinations (5 radii × 5 speeds) of the full factorial test into 15 key combinations. This reduces redundant testing while ensuring the statistical analysis of the data. The specific scheme is shown in Table 1.

[0037] Table 1. Speed ​​Test Combination Schemes for Circular Curves with Different Radius

[0038] By concentrating resources on high-risk, small-radius intervals and critical speed conditions, the above approach helps to capture abrupt changes in physiological indicators such as electrocardiogram, skin conductance, and electromyography, providing data support for subsequent derivation of speed limits based on the driver's physiological load.

[0039] Multiple physiological indicators were collected: Heart rate is a core indicator for measuring the activity of the human cardiovascular system. The normal resting heart rate range for adults is 60-100 beats per minute. When an individual faces stress or a heavy workload, the sympathetic nervous system is activated, and the heart rate rises accordingly. Driving on small-radius circular curves on mountain roads is challenging, and drivers experience corresponding fluctuations in heart rate when navigating these curves. These fluctuations can, to some extent, characterize the intensity of the driving load.

[0040] This embodiment uses the ECG100C ECG amplifier of the BIOPAC system to acquire the driver's ECG signal, and the sampling frequency is set to 1000Hz to meet the requirements of high-fidelity ECG signal acquisition.

[0041] Because there are significant differences in baseline heart rate among individuals, it is difficult to conduct cross-sectional comparisons and standardized assessments using absolute heart rate values ​​directly. Therefore, this implementation method uses the heart rate growth rate as the evaluation index. The heart rate growth rate standardizes heart rate changes into a percentage form, effectively eliminating the interference of individual differences and making the physiological load of different drivers under the same driving conditions comparable. ; in, Heart rate growth rate, % The average heart rate, in bpm, is for the curved road section. Resting reference heart rate, bpm.

[0042] When drivers face challenging driving tasks, sympathetic nerve activation leads to increased sweat gland secretion, resulting in elevated skin conductivity. During driving on small-radius circular curves, the increased complexity of maneuvers and risk perception cause significant changes in the driver's skin conductivity, reflecting their level of psychological stress and emotional state.

[0043] Data acquisition was performed using the EDA100C skin conductivity amplifier from the BIOPAC system, with the sampling frequency set to 125Hz.

[0044] To eliminate individual differences in baseline skin conductance levels, this implementation method uses the skin conductance growth rate as the evaluation index. The skin conductance growth rate standardizes skin conductance changes into a percentage, ensuring comparability of data from different individuals and under different experimental conditions. ; in, The skin cell growth rate was % The average skin charge value, in μS, is the value of the circular curve section. The resting reference skin conductance value is given in μS.

[0045] Electromyography (EMG) signals can directly characterize the intensity of muscle activity during a driver's steering maneuvers. When driving on small-radius circular curves, the driver needs to continuously adjust the steering wheel angle, which increases the activation level of related muscle groups. By monitoring the EMG activity of these muscle groups, the driver's workload under different road conditions can be quantitatively assessed.

[0046] This embodiment focuses on monitoring the electromyographic activity of the flexor and extensor muscles of the driver's right forearm. The EMG100C electromyographic amplifier of the BIOPAC system is used for signal acquisition, and the sampling frequency is set to 2000Hz.

[0047] This implementation method adopts As an evaluation index of electromyographic signals. The value can characterize the actual level of muscle exertion. This index is sensitive to fluctuations in signal amplitude and is suitable for quantifying muscle activity intensity. Calculations based on the square property of the signal can handle the problem of alternating positive and negative values ​​in electromyographic signals, resulting in more stable measurement results. ; in, The root mean square (RMS) value represents the effective amplitude of the signal. This refers to the total number of samples or sampling points within the sampling time window. For the first The original electromyography signal amplitude at each sampling point is typically expressed in microvolts (μV) or millivolts (mV). This is the index of the sample sequence.

[0048] After the raw electromyography (EMG) signals were acquired, AcqKnowledge software was used for data processing. First, adaptive filtering was used to remove interference signals, and then calculations were performed. After processing, the data is exported for subsequent analysis.

[0049] This implementation method selects three physiological indicators—heart rate, skin conductance, and electromyography (EMG)—for joint analysis. The ECG indicator is used to capture the stress response of the cardiovascular system, the skin conductance indicator focuses on characterizing the emotional arousal state of the neuroendocrine system, and the EMG indicator reflects the operational load of the motor system. While these three indicators differ in their temporal characteristics and sensitivity, they are related in their physiological mechanisms and can reflect the driver's overall workload level from different dimensions.

[0050] Multiple physiological indicators need to be synchronized during the data acquisition process, and data quality also needs to be controlled. The BIOPAC system uniformly controls the acquisition sequence of all physiological signal acquisition modules, ensuring precise alignment of the ECG, ductal electrocardiogram (ECG), and electromyography (EMG) signals on the time axis. A time synchronization mechanism is also established between the driving simulator system and the physiological data acquisition system, enabling frame-by-frame matching of vehicle position information and physiological data.

[0051] During the acquisition process, the quality status of each physiological signal was monitored in real time using AcqKnowledge software, and signal quality evaluation standards were established: the signal-to-noise ratio of the ECG signal was greater than 20dB, and the R wave was clearly distinguishable; the baseline of the ductus skin tissue signal was stable with no obvious drift; and the amplitude of the electromyography (EMG) signal was less than 10μV at rest. Corresponding preprocessing schemes were developed based on the characteristics of different types of physiological signals: the ECG signal was filtered using a 0.5~35Hz bandpass filter to remove baseline drift and high-frequency noise, and the Pan-Tompkins algorithm was used for R-wave detection; the ductus skin tissue signal was filtered using a 0.01~5Hz low-pass filter to smooth the signal and remove motion artifacts; and the EMG signal was filtered using a 20~450Hz bandpass filter and subjected to full-wave rectification.

[0052] Based on the aforementioned speed-radius combination scheme, each driver needs to complete 12 driving tasks. To avoid the effects of fatigue and learning, a Latin square balanced design was used to determine the order of task execution. The time sequence of each driving task included: baseline data collection (3 minutes), during which the subject sat quietly in the driver's seat in a relaxed state; road familiarization (2 minutes), during which the driver slowly passed through the test road section at a speed of 30 km / h; formal driving task (5 minutes), during which the subject completed a circular curve driving task at a specified speed; and recovery period (3 minutes), during which the driver rested to allow physiological indicators to return to baseline levels.

[0053] The environmental conditions in the data collection room were strictly controlled: the room temperature was maintained at 22±2℃, and the background noise was less than 40dB. The illuminance inside the driving simulator was set to 300~500 lux to simulate daytime driving conditions. During the data collection process, the subject's condition was monitored in real time by observing the quality of physiological signals. After every three driving tasks, a simplified fatigue scale was used to assess the subject's subjective fatigue level. At the same time, vehicle trajectory deviation and speed control accuracy were recorded to ensure that the subject completed the driving tasks seriously.

[0054] During data acquisition, the BIOPAC system recorded and saved the driver's electrocardiogram (ECG), electrical activity of the skin, and electromyography (EMG) signals in real time. After the acquired data was imported into the AcqKnowledge software, the software could display time-domain waveforms of various indicators. After analysis and processing, the data could be exported in TXT or Excel format for subsequent statistical analysis and calculation. After acquisition, the physiological data was exported to multiple formats using AcqKnowledge 5.0 software, including the original .acq format file for further in-depth analysis, and .txt and .xlsx format files for statistical software processing. The exported data included timestamps, raw signal values ​​for each channel, and preprocessed feature parameters.

[0055] This implementation method establishes a multi-level data quality inspection mechanism: integrity inspection ensures that the data collection time for each collection condition is no less than 5 minutes and the number of sampling points meets the expected requirements; consistency inspection uses correlation analysis to check the consistency of baseline data of the same subject at different time periods, and the correlation coefficient should be greater than 0.85; outlier detection adopts the 2σ criterion, combined with physiological rationality judgment, and marks and processes data points that deviate significantly from the normal range. The collected raw data are shown in Table 2.

[0056] Table 2. Raw physiological data (partial)

[0057] Step 2: After processing the collected multiple physiological indicators, calculate the reliability and individual variability of different physiological indicators, and statistically analyze the safety thresholds corresponding to different physiological indicators. This implementation method collected physiological data from five drivers under different circular curve radii and driving speeds, including three core physiological indicators: heart rate, skin conductance, and electromyography. To ensure data comparability and the accuracy of subsequent analysis, the raw data were first systematically preprocessed and standardized.

[0058] Raw physiological data were recorded using the BIOPAC system at a sampling rate of 1000 Hz. Each driver completed 15 trials, resulting in 75 sets of valid data. Data preprocessing included three steps: signal filtering, baseline correction, and feature extraction. Electrocardiogram (ECG) signals were filtered using a 0.5–40 Hz bandpass filter to remove high-frequency noise and baseline drift, and heart rate values ​​were extracted using an R-wave detection algorithm. Electrodermal activity (EDA) signals were filtered using a 0.05 Hz high-pass filter to eliminate slow drift, and the increment value relative to the baseline was calculated. Electromyography (EMG) signals were filtered using a 20–500 Hz bandpass filter, and the root mean square (RMS) value was calculated as a quantitative indicator of muscle activity intensity.

[0059] In this embodiment, the statistical data of different drivers in terms of individual characteristics and physiological baseline are shown in Table 3.

[0060] Table 3. Statistics on individual characteristics and physiological indicators of drivers

[0061] Table 3 shows that there are significant differences in the baseline levels of physiological indicators among different drivers. Driver 4, with 15 years of driving experience, had the lowest mean and smallest standard deviation for all physiological indicators (heart rate 69.1±1.9 bpm, skin conductance 0.77±0.24 μS, electromyography 0.009±0.002 mV), reflecting strong physiological regulation ability and rich driving experience. Driver 3, with 1 year of driving experience, showed the highest level of physiological arousal and greater fluctuation (heart rate 88.7±3.2 bpm, skin conductance 2.41±0.47 μS, electromyography 0.024±0.005 mV), reflecting the high sensitivity characteristics of novice drivers.

[0062] Therefore, to eliminate differences in baseline physiological levels among individual drivers, this implementation method uses relative growth rates to standardize physiological indicators. The heart rate growth rate and skin conductance growth rate represent the percentage change in heart rate and skin conductance relative to a baseline state during driving, effectively eliminating individual differences and making the physiological load of different drivers comparable under the same driving conditions. For electromyography (EMG) signals, the obtained EMG RMS values ​​after individual standardization are directly used for subsequent analysis.

[0063] Figure 2 Using heart rate as an example, we can illustrate the individual differences among drivers and the characteristics of data distribution. Figure 2(a) shows the heart rate distribution of 5 drivers under 15 test conditions. Driver 3 had the highest mean (88.5 bpm) and driver 4 had the lowest mean (69.1 bpm), reflecting significant differences in individual physiological baseline levels. Figure 2 (b) shows the data distribution of each driver after standardization by growth rate. Driver 3 has the largest fluctuation range (13.5%-28.4%), while driver 4 has the smallest (3.1%-14.1%), indicating that novices are more sensitive to physiological responses to driving tasks, while experienced drivers have stronger physiological regulation capabilities. Figure 2 (c) The data shows a right-skewed distribution across all 75 data sets. Statistical analysis revealed a median of 12%, a 75th percentile of 18%, and a 90th percentile of 23%. 90% of the data were below 23%, with only a few high-difficulty driving conditions exceeding this level.

[0064] Outlier detection was performed on the standardized data. The 2σ criterion was used, where x < μ - 2σ or x > μ + 2σ was considered a statistical outlier, where x is a single sample observation, μ is the sample mean, and σ is the sample standard deviation. Combined with physiological rationality assessment, only 7 statistical outliers were identified in the 75 data sets, including 2 in heart rate, 3 in skin conductance, and 2 in electromyography (EMG), accounting for 9.3% of the total data. These outliers mainly came from driver 3 with 1 year of driving experience, and occurred in stress responses under challenging driving conditions. All outliers were within physiologically reasonable ranges: heart rate below 150 bpm, skin conductance below 10 μS, and EMG below 0.5 mV, reflecting the true physiological state of novice drivers facing challenging driving tasks. Considering the important reference value of these data in determining safe speed limits, it was decided to retain them. These outliers represent a small percentage and are all within physiologically reasonable ranges, and will not have a significant impact on subsequent statistical analysis and speed limit determination.

[0065] This implementation method comprehensively evaluates data quality from three dimensions: measurement stability, inter-individual variability, and indicator consistency.

[0066] First, the measurement stability of each physiological indicator was assessed using the intraclass correlation coefficient (ICC). The ICC calculation formula is as follows: ; in, For the mean square between groups, Within-group mean square This represents the number of times the measurement was repeated.

[0067] In this embodiment, the reliability evaluation criteria for ICC values ​​are as follows: ICC < 0.50 indicates poor reliability, 0.50 ≤ ICC < 0.75 indicates moderate reliability, 0.75 ≤ ICC < 0.90 indicates good reliability, and ICC ≥ 0.90 indicates excellent reliability. The reliability test results of the three physiological indicators are shown in Table 4.

[0068] Table 4. Results of Reliability Tests for Physiological Indicators

[0069] As can be seen from the reliability test results in Table 4, the measurement stability of the three physiological indicators exhibits a clear hierarchical characteristic. In this embodiment, the ICC values ​​of all three physiological indicators exceed 0.60, reaching a moderate to high level, meeting the basic requirements for statistical analysis. The heart rate indicator shows the highest measurement stability (ICC=0.877), reaching a good level, and its confidence interval is relatively narrow [0.571, 0.985], indicating that it has a stable physiological regulatory mechanism and strong anti-interference ability as a circulatory system indicator. The reliability of the electromyography (EMG) indicator is moderately high (ICC=0.784). This moderate variability actually reflects the differences in muscle exertion habits among different drivers and the cumulative effect of muscle fatigue, which is within an acceptable range and physiologically reasonable. The ICC value of the electrodermal skin (EDS) indicator is 0.608, reaching a moderate level. As an indicator reflecting the activity of the sympathetic nervous system, EDS can sensitively capture changes in the driver's emotional stress response and is an important dimension for evaluating the driver's psychological tension.

[0070] Overall data quality evaluation: The ICC values ​​of all three indicators exceeded 0.60, indicating that the overall measurement quality reached an acceptable level. The high stability of the heart rate indicator makes it suitable as a core evaluation indicator; the moderately high reliability of the electromyography indicator supports its use as a tool for measuring operational load; and although the electrodermal conductivity indicator has greater variability, its sensitivity makes it uniquely valuable in capturing changes in emotional stress.

[0071] Secondly, the magnitude of individual differences was assessed by calculating the coefficient of variation (CV) among drivers under different test conditions. The coefficient of variation reflects the dispersion of physiological responses among different drivers under the same driving conditions. ; in, and The first The standard deviation and mean of the five drivers under the given conditions.

[0072] The coefficient of variation (CV) eliminates the influence of units by using the ratio of standard deviation to mean, and can directly compare the degree of variation of different physiological indicators. In this embodiment, CV < 10% indicates that the degree of variation is very small and the data is highly stable; 10% ≤ CV < 15% indicates that the degree of variation is small and the data is relatively stable; 15% ≤ CV < 30% indicates that the degree of variation is moderate; and CV ≥ 30% indicates that the degree of variation is large and the data is unstable.

[0073] Figure 3 This demonstrates the degree of difference in physiological responses among individuals under different conditions. The first number on the horizontal axis label represents the radius of the circular curve (m), and the second number represents the driving speed (km / h). For example, "60-30" indicates a test condition with a radius of 60m and a speed of 30km / h. The CV value of the heart rate index remained at a low level of 8.2-17.3%, with an average of 12.4%, showing good inter-individual consistency. The CV value of the skin conductance index ranged from 18.2-35.6%, with an average of 28.5%, and the CV value of the electromyography index ranged from 16.8-32.4%, with an average of 27.8%, both showing moderate variability overall. Although the CV values ​​of skin conductance and electromyography were slightly higher than 30% under challenging driving conditions such as 70-45 and 80-50, they remained below 30% for the vast majority of conditions, reflecting individual differences in emotional responses and driving habits.

[0074] Coefficient of variation analysis showed that the heart rate index had the best inter-individual consistency, with a coefficient of variation of 12.4%. The skin conductance and electromyography indices showed moderate inter-individual differences, with coefficients of variation of 28.5% and 27.8%, respectively, which were at a moderate level of variation.

[0075] This implementation method performs statistical analysis on multiple collected physiological indicators to determine the safety threshold for each physiological indicator.

[0076] Heart rate increase rate directly reflects the stress intensity of driving tasks on the circulatory system. Statistical results from 75 samples in this embodiment show a median heart rate increase rate of 12%, a 75th percentile of 18%, and a 90th percentile of 23%. A sustained increase in heart rate increases myocardial oxygen consumption, and maintaining a high level for extended periods can easily lead to fatigue accumulation and decreased attention. Therefore, this embodiment sets a heart rate increase rate of 20% as a safe threshold for stress response. This value falls between the 75th and 90th percentiles, providing a safety margin for most drivers without threatening driving safety due to excessive activation of the cardiovascular system.

[0077] Skin conductance activity is regulated by the sympathetic nervous system, and its amplitude is closely related to the degree of emotional arousal. Statistical results show that the distribution of skin conductance growth rate data is right-skewed, with a median of 21.8%, a 75th percentile of 27.2%, and a 90th percentile of 31.4%. Considering the synergistic relationship between skin conductance indicators and heart rate and electromyography, this implementation method sets a skin conductance growth rate of 30% as an acceptable upper limit for emotional stress. This threshold is slightly higher than the 75th percentile but lower than the 90th percentile, allowing for some leeway for individual differences while ensuring the safety of most drivers.

[0078] The intensity of electromyographic (EMG) activity in the forearm muscles reflects the driver's load level during steering. Statistical results show that the median EMG RMS value is 0.0155 mV, the 75th percentile is 0.019 mV, and the 90th percentile is 0.0215 mV. When EMG activity remains at a high level for an extended period, muscle fatigue accumulation accelerates significantly. Based on this, this embodiment sets the EMG RMS value of 0.022 mV as the safe threshold for operational load. This value is close to the 90th percentile, ensuring that the vast majority of drivers are in a relatively comfortable operating state.

[0079] Step 3: Based on the safety thresholds, reliability, and individual variability corresponding to different physiological indicators, construct a comprehensive physiological load index model.

[0080] A single physiological indicator is insufficient to fully reflect the driver's overall workload status. Therefore, this implementation method constructs a comprehensive physiological workload index model: ; in, For heart rate growth rate, For skin cell growth rate, This is the root mean square value of electromyography. This is a comprehensive physiological load index.

[0081] In this embodiment, the weight allocation comprehensively considers the measurement reliability and inter-individual variability of each indicator. The heart rate indicator has the highest ICC value and the lowest coefficient of variation, showing the best measurement stability, and is given the maximum weight of 0.5; the electromyography indicator has moderate to high reliability and is given a weight of 0.3; the electrodermal conductivity indicator is sensitive to emotional stress response and can capture changes in the driver's psychological tension level, and is given a weight of 0.2.

[0082] Implementation Method 2: A method for determining speed limits on small-radius circular curves in mountainous areas. This method is based on the comprehensive physiological load index model construction method for determining speed limits described in Implementation Method 1, and includes the following steps: Step A1: For different radii of circular curves on mountain roads, preset multiple vehicle speeds respectively; Step A2: Obtain multiple physiological indicators of the driver when passing through different circular curve radii on mountain roads at different vehicle speeds; Step A3: Use the multiple physiological indicators obtained in step A2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value for the radius of the corresponding circular curve of the mountain road based on the comprehensive physiological load index.

[0083] In step A1, multiple vehicle operating speeds are preset for different circular curve radii on mountain roads, including low-speed, medium-speed, and high-speed conditions, specifically: When the radius of the circular curve on the mountain road is 60m, the preset vehicle speeds are 30km / h, 35km / h and 40km / h. When the radius of the circular curve on the mountain road is 70m, the preset vehicle speeds are 35km / h, 40km / h and 45km / h. When the radius of the circular curve on the mountain road is 80m, the preset vehicle speeds are 40km / h, 45km / h and 50km / h. When the radius of the circular curve on the mountain road is 90m and 100m, the preset vehicle speed is 45km / h, 50km / h and 55km / h.

[0084] Using the same multiple physiological index collection method as in Implementation Method 1, multiple physiological indexes of the driver were collected when passing through different circular curve radii of mountain roads at different vehicle operating speeds as preset above.

[0085] The acquired multiple physiological indicators are used as inputs to the comprehensive physiological workload index (PLI) model to calculate the comprehensive physiological workload index. Based on the characteristics of the PLI index construction, this implementation method divides the driver's comprehensive physiological workload index (PLI) into four levels: Light load (PLI<0.60): The driver is in a relaxed state with a low level of physiological arousal, and road resources are not fully utilized.

[0086] Moderate load (0.60≤PLI<0.80): The driver is in a moderate state of alertness, the physiological system is appropriately activated but not over-stressed, and can maintain a good level of attention while having sufficient coping reserves.

[0087] High load (0.80≤PLI<1.0): The driver enters a state of high tension, the physiological system approaches the safety boundary, and the reserve capacity to deal with emergencies is reduced. This state can be tolerated for a short time, but maintaining it for a long time will lead to rapid accumulation of fatigue and is not suitable as a regular driving state.

[0088] Overload (PLI ≥ 1.0): The driver's physiological system exceeds the safety threshold, with multiple physiological indicators simultaneously at high levels, posing a significant safety risk. A PLI of 1.0 indicates that at least one physiological indicator has reached the safety threshold.

[0089] This implementation systematically evaluates three test speeds for each radius and determines the speed limit values ​​for circular curves of different radii: Determining speed limits requires comprehensive consideration of driver physiological safety and road traffic efficiency. This implementation method employs a direct evaluation approach based on measured data. By systematically analyzing the distribution characteristics of the Comprehensive Physiological Load Index (PLI) of driver groups at three test speeds using circular curves of different radii, it identifies the optimal speed value that satisfies both safety requirements and maximizes traffic efficiency.

[0090] Determining speed limits requires meeting two basic principles: First, the physiological safety principle, which requires that the PLI of most drivers at that speed does not exceed the safe threshold, ensuring that the physiological system is not under excessive stress; second, the operational feasibility principle, which requires that the speed limit be within the actual operating speed distribution range, neither too conservative, leading to low traffic efficiency, nor too aggressive, deviating from the characteristics of actual driving behavior.

[0091] This implementation sets three test speeds—low, medium, and high—for each radius circular curve. By comparing the PLI distribution characteristics of the driver group under these three speed conditions, the optimal speed value that meets safety criteria is identified. This method is based on actual measured physiological data, avoiding the bias of theoretical predictions, while fully considering the real impact of individual differences.

[0092] The specific judgment criteria include the following aspects: First, the average PLI should be in the moderate load range (0.60~0.80). This range represents the ideal state where the driver is moderately alert rather than highly tense. By comparing the three speed levels of low speed, medium speed and high speed corresponding to the radius of each circular curve, it was found that: (1) Low speed condition: average PLI 0.44~0.67, mostly in the light load range, although the physiological pressure is low, the road utilization rate is low; (2) Medium speed condition (recommended speed limit): average PLI 0.70~0.77, in the middle and upper position of the moderate load range, achieving the best balance between physiological safety and road efficiency; (3) High speed condition: average PLI 0.82~0.87, entering the high load range, the driver's physiological system is close to the safety boundary.

[0093] It should be noted that for a typical minimum radius alignment of 100m, the average PLI (Power Index) at high speed (55km / h) is 0.70, still within the moderate load range, and all drivers' individual physiological indicators did not exceed the safety threshold. This indicates that good geometry significantly reduces driving difficulty, allowing this radius to maintain a moderate load at higher speeds. Therefore, a speed limit of 55km / h is recommended for a 100m radius.

[0094] For a radius of 60-90m, the average PLI (Power Index) under high-speed conditions is no less than 0.82, which is already in the high-load range and is not suitable as a recommended speed limit. Although these values ​​have not exceeded the physiological limit of 1.0, they have already put the driver in a state of high tension, exposing them to the following risks: (1) Insufficient physiological reserves to cope with emergencies. In actual roads, there may be emergencies such as oncoming vehicles, pedestrians crossing, and road obstacles, which require additional physiological resources to respond to stress. If the baseline load has reached 0.85, any additional stress may cause the PLI to exceed the physiological limit of 1.0; (2) The rate of fatigue accumulation is accelerated. The duration of a single test in this embodiment is only 5 to 6 minutes, while actual driving on mountain roads may last for tens of minutes or even several hours. Long-term operation under high load conditions will lead to rapid accumulation of fatigue and affect driving safety; (3) Individual differences lead to overloading by some drivers. Although the average PLI under high-speed conditions is 0.82~0.87, the individual PLI of novice drivers is significantly higher than the average level, and they have approached or entered a state of physiological overload under some challenging conditions.

[0095] Therefore, the core criterion for recommending a speed limit is that the average PLI should be maintained within the moderate load range (0.60~0.80). For a radius of 60~90m, a medium speed condition is selected, with an average PLI of 0.70~0.77; for a radius of 100m, a high speed condition of 55km / h is selected, with an average PLI of 0.70, which also meets this criterion, so this speed is recommended.

[0096] Second, the percentage of individual physiological indicators exceeding the threshold should be used as supplementary verification. Although PLI is a comprehensive evaluation indicator, the independent response characteristics of each physiological system need to be considered. Under medium-speed conditions, the percentage of individual indicators exceeding the threshold is 4-8%, mainly due to the stress response of novice drivers under certain high-difficulty combinations. The vast majority of drivers (>90%) have all physiological systems within a safe range. In contrast, under high-speed conditions, the percentage of heart rate exceeding the threshold reaches 32%, and the percentage of electromyography exceeding the threshold reaches 28%, further validating the rationality of the medium-speed conditions.

[0097] like Figure 4As shown, although the safety margin is sufficient under low-speed conditions, the average PLI is too low (0.44~0.67), and road resources are not fully utilized; under high-speed conditions, the average PLI of most radii exceeds 0.80 and enters the high-load range; under medium-speed conditions (and 100m high-speed conditions), the average PLI is maintained in the moderate load range of 0.70~0.77, achieving the best balance, and therefore becomes the recommended speed limit value for each radius.

[0098] Figure 4 The distribution characteristics of PLI at three test speeds are shown in the form of a box plot. The horizontal axis represents the radius of the circular curve, and the vertical axis represents the comprehensive physiological load index (PLI).

[0099] from Figure 4 The data distribution reveals a clear stratification across the three test speeds for each radius condition. Low-speed conditions, represented by blue, have a mean PLI between 0.51 and 0.58, falling within the light load range. The majority of the test cases are located below the moderate load range (0.60–0.80), with all upper quartiles below 0.65, indicating that all drivers have sufficient physiological safety margins at this speed. However, the excessively low mean PLI also suggests that road resources are not being fully utilized, resulting in a loss of traffic efficiency.

[0100] Medium-speed conditions are represented in green. The average PLI for a radius of 60-90m ranges from 0.73 to 0.77, placing the main body of the vehicle within the moderate load range (0.60-0.80). The upper quartile is between 0.75 and 0.79, with a few upper limits slightly exceeding 0.80 but still well below physiological limits. This distribution indicates that the physiological load of most drivers is within a reasonable range, while significantly improving road utilization efficiency compared to low-speed conditions. The average PLI for medium-speed conditions with a 60-meter extreme radius reaches 0.76, close to the upper limit of the moderate load range, reflecting the driving difficulty under extreme radius conditions. This suggests that the 35km / h speed limit is a reasonable upper limit for this radius.

[0101] High-speed conditions are indicated in red. The average PLI (Power Index) for a radius of 60-90m ranges from 0.82 to 0.87. The main part of the box exceeds the upper limit of the moderate load range (0.80), entering the high load range. This indicates that this speed condition has put drivers in a state of high stress, with some drivers approaching their physiological limits. The box plot clearly shows that under high-speed conditions with radii of 60m, 70m, and 80m, the PLI box positions are significantly higher and the dispersion is increased. This indicates that under these dangerous combinations of small radii and high speeds, the physiological responses of drivers are significantly differentiated, with some drivers entering a state of over-stress. The situation is different for high-speed conditions with a radius of 100m (55km / h). Its average PLI is approximately 0.70, still within the moderate load range. This is the data basis for recommending a speed limit of 55km / h for a 100m radius.

[0102] The formation mechanism of this stratified distribution can be explained from the perspective of driver physiological responses. At low speeds, lateral acceleration is small. Although the driver needs to turn the steering wheel to navigate curves, the operation is not difficult, muscle exertion is light, and psychological pressure is low, resulting in a low overall physiological load. At medium speeds, lateral acceleration increases to a moderate level. The driver needs to actively adjust the steering wheel angle and apply some muscle force, leading to a corresponding increase in heart rate and skin electrical activity. However, this increase remains within a tolerable range, creating a "vigilant but not tense" driving state. At high speeds, lateral acceleration further increases. The driver needs to maintain a high level of attention and apply greater muscle force to control the vehicle, significantly increasing psychological tension and consequently, a significant rise in overall physiological load. This pattern is observed on circular curves of different radii, but on smaller radius curves, due to sharper curves and shorter visibility, even at the same speed, the driver experiences greater psychological pressure and operational difficulty, leading to a more pronounced stratification of physiological load.

[0103] Through the Figure 4 A lateral comparison of the five radius conditions reveals that, under recommended speed limits for curves with radii ranging from 60 meters to 100 meters, the mean PLI (Power Index) for each radius remains within the moderate load range of 0.70 to 0.77, exhibiting a fairly consistent distribution. This demonstrates that a differentiated speed limit strategy can maintain a similar moderate load level for drivers under different radius curve conditions, ensuring driving safety while avoiding the waste of road resources caused by a uniform speed limit.

[0104] Based on the above judgment criteria and Figure 4 Based on the PLI distribution characteristics shown and the safety threshold standards determined in Implementation Method 1, this implementation method identifies recommended speed limits for five circular curves with different radii. Table 5 summarizes the recommended speed limits and their key verification indicators.

[0105] Table 5 Recommended Speed ​​Limits for Small-Radius Circular Curves on Mountain Roads

[0106] In Table 5, "over-threshold ratio" refers to the proportion of drivers with a PLI > 0.80 at this speed; "difference from design speed" refers to the deviation from the design speed of 40 km / h for a Class III mountain highway.

[0107] (1) General characteristics of recommended speed limits Table 5 shows that the average PLI corresponding to the five recommended speed limits remains within the moderate load range (0.60-0.80) of 0.70-0.77. Except for one driver with a PLI slightly exceeding 0.80 at an 80m radius, the PLI of all drivers at other radii remains below 0.80. By matching corresponding speed values ​​to different radii, drivers maintain a similar physiological load level under different curve conditions. The lateral acceleration generated by the recommended speed limits is 1.58-2.33 m / s², all lower than the comfort driving upper limit of 2.5 m / s².

[0108] The recommended speed limit increases with radius, in increments of 5 km / h. At 70 meters, it matches the design speed of 40 km / h, decreasing to 35 km / h at 60 meters, and gradually increasing to 45-55 km / h between 80 and 100 meters. This differentiated speed limit setting reflects the core concept of this implementation method: reducing speed limits in small-radius sections to ensure safety, and increasing speed limits in large-radius sections to improve efficiency.

[0109] (2) Detailed analysis of recommended speed limits for each radius The recommended speed limit for a 60-meter radius is 35 km / h. Even at this speed limit, the average PLI (Power Index) reaches 0.76, close to the upper limit of the moderate load range. The lateral acceleration generated by this speed limit is 1.58 m / s², providing sufficient physical safety margin for this radius condition.

[0110] The recommended speed limit for a 70-meter radius is 40 km / h, with an average PLI of 0.75, placing it in the upper-middle range of moderate load. This result, from a data perspective, verifies the rationality of the current design standard under a 70-meter radius condition and also demonstrates that 70 meters can serve as a key transition point from extreme to general load conditions.

[0111] The recommended speed limit for an 80-meter radius is 45 km / h, a 12.5% ​​increase over the design speed, with an average PLI of 0.77, close to the upper limit of the moderate load range. This increase reflects the improved speed adaptability of drivers as the radius increases and road geometry improves, but drivers still need to stay within safe limits.

[0112] The recommended speed limit within a 90-meter radius is 50 km / h, 25% higher than the design speed, with an average PLI of 0.73, placing it in the middle of the moderate load range. This speed limit produces a lateral acceleration of 2.14 m / s², within the range for comfortable driving.

[0113] The recommended speed limit within a 100-meter radius is 55 km / h, 37.5% higher than the design speed, with an average PLI of 0.70, placing it in the lower-middle range of moderate driving load. The favorable alignment ensures that this radius maintains a moderate physiological load even at higher speeds. The lateral acceleration generated by this speed limit is 2.33 m / s², lower than the comfort driving limit of 2.5 m / s², validating the positive effect of good alignment on reducing driving difficulty.

[0114] (3) Relationship between lateral acceleration and physiological load Lateral acceleration only reflects the physical motion of the vehicle and cannot fully represent the driver's actual physiological experience. As can be seen from Table 5, although the lateral acceleration at a radius of 100 meters and a speed of 55 km / h is 2.33 m / s², which is significantly higher than the 1.58 m / s² at a radius of 60 meters and a speed of 35 km / h, the former has a PLI of 0.70, which is significantly lower than the latter's 0.76.

[0115] This phenomenon reveals the complex mechanism by which small-radius circular curves affect the driver's physiological load. Although the lateral acceleration value is relatively small at a radius of 60 meters, the small radius, sharp curve, and short visibility distance mean that drivers face greater psychological pressure and operational challenges when navigating curves: they need to adjust the steering wheel angle more frequently, control the vehicle trajectory more precisely, and focus more intently on judging the relationship between the vehicle's position and the road boundary. These additional cognitive and operational loads lead to a significant increase in electromyographic activity and a marked rise in psychological stress levels, ultimately resulting in a relatively high PLI of 0.76.

[0116] Therefore, the danger of small-radius curves stems not only from physical lateral forces but also from the high demands placed on the driver's psychological and operational abilities by the curve's geometry. Although the lateral acceleration of 1.58 m / s² is not particularly high, the visual pressure and operational tension created by the small radius significantly increase the driver's overall physiological load. In contrast, while the lateral acceleration of a 100-meter radius reaches 2.33 m / s², the curve is gentler and visibility is better, resulting in lower psychological pressure and operational difficulty for the driver, and the overall physiological load remains at a moderate level of 0.70.

[0117] To further illustrate the limitations of traditional methods, compare the performance of a 60-meter radius at different speeds: when the speed increases from 35 km / h to 40 km / h, the lateral acceleration increases from 1.58 m / s² to 2.06 m / s², still far below the comfort upper limit of 2.5 m / s². From a physical perspective, it should be safe. However, the test data shows that at this time, 80% of the drivers have at least one physiological index exceeding the safety threshold. This indicates that the traditional speed limit method based on lateral acceleration only considers the physical performance of the vehicle and ignores the psychological perception and physiological tolerance of the driver.

[0118] (4)Safety margin analysis The PLI of larger radii (90m, 100m) is relatively low (0.70 - 0.73), with a large margin from the upper limit (0.80) of the moderate load range, which can provide a buffer space for adverse conditions such as bad weather and road condition degradation. The PLI of smaller radii (60m, 70m, 80m) is relatively high (0.75 - 0.77), but still remains within the moderate load range, ensuring basic safety.

[0119] For the transition values (such as 65m, 73m, 85m, etc.) of the radius R ∈ [60, 100], the recommended speed limit value v can be determined by linear interpolation. Let R be between the standard radii R1 and R2 (R1 < R < R2), and the corresponding speed limit values be v1 and v2, then: ; The calculation result is rounded to a multiple of 5 km / h. For example, for a 73-meter radius: ; After rounding, it is 40 km / h. Through this linear interpolation method, the speed limit value of the transition radius can be determined without conducting specialized tests for each radius.

[0120] In summary, this embodiment determines the recommended speed limit values for five standard radii based on the driver's physiological load. Compared with the traditional speed limit method based on lateral acceleration, this embodiment further considers the physiological response characteristics of the driver by introducing physiological indicators such as heart rate, skin conductance, and electromyogram, starting from the actual tolerance of the driver and ensuring the physical safety of the vehicle.

[0121] Embodiment 3. Method for determining the speed limit value of a small-radius circular curve on a mountain road. The method is implemented based on the comprehensive physiological load index model construction method for speed limit value determination described in Embodiment 1, and includes the following steps: Step B1, obtain the radius of the circular curve of the mountain road and preset different vehicle operating speeds; Step B2: Construct a physiological indicator prediction model and calculate the corresponding physiological parameters based on the radius of the circular curve and different vehicle operating speeds; Step B3: Use the physiological parameters calculated in step B2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value corresponding to the radius of the circular curve of the mountain road based on the comprehensive physiological load index.

[0122] Existing technologies do not consider the impact of the radius of the circular curve and the vehicle speed on the driver's physiological indicators. Moreover, the speed limit value determined based on the driver's physiological indicators usually relies on driving simulation or real vehicle tests to collect actual physiological data under each combination of radius and speed. This has shortcomings such as high test costs, long evaluation cycle, only covering a limited number of discrete test points, and difficulty in quickly extrapolating to untested road sections.

[0123] To address the aforementioned technical issues, this implementation method constructs a physiological index prediction model that can calculate physiological parameters and the comprehensive physiological load index using only the radius of the circular curve and a preset speed. This eliminates the need for repeated driving tests, significantly shortens the evaluation cycle, reduces implementation costs, and supports continuous prediction of transition radii and non-standard speed conditions. It is also easy to integrate into road design software or traffic management systems to achieve rapid evaluation and dynamic updating of speed limits for large-scale road networks.

[0124] In this embodiment, the physiological indicator prediction model in step B2 includes a heart rate growth rate prediction model, a skin conductance growth rate prediction model, and an electromyography root mean square value prediction model.

[0125] The heart rate growth rate prediction model is as follows: ; The skin cell growth rate prediction model is as follows: ; The root mean square value prediction model for electromyography is as follows: ; in, The radius of the circular curve is... The vehicle's operating speed.

[0126] The calculated physiological indicators are used as input to the comprehensive physiological load index model to calculate the comprehensive physiological load index. The comprehensive physiological load index is divided into multiple interval levels, and the highest vehicle operating speed that makes the comprehensive physiological load index fall within the moderate load range is selected as the recommended speed limit value for the radius of the corresponding mountain road circular curve.

[0127] This implementation method achieves rapid cross-domain prediction of factors ranging from road geometry parameters to driver physiological load by constructing a physiological index prediction model. Compared to actual measurement methods, it avoids repeatedly conducting driving simulations or real-vehicle tests for each new road section, significantly reducing the time and economic costs of determining speed limits. Furthermore, it accurately reflects the impact of factors such as limited visibility and high operational complexity in small-radius curves on driver physiology, meeting the needs of determining speed limits for mountain roads. This ensures that the determined speed limits meet vehicle mechanical safety requirements while remaining within the driver's physiological tolerance threshold, significantly improving the operational safety of mountain roads.

[0128] Implementation Method 4: This implementation method compares the method of directly evaluating and determining the recommended speed limit using measured data in Implementation Method 2 with the method of determining the recommended speed limit by constructing a physiological index prediction model as described in Implementation Method 3. The two methods mutually verify each other's effectiveness.

[0129] The heart rate growth rate, skin conductance growth rate, and root mean square value of electromyography calculated by the physiological index prediction model established in Implementation Method 3 are substituted into the comprehensive physiological load index model.

[0130] By substituting the recommended speed limits for each radius into the aforementioned physiological index prediction model, the corresponding theoretical PLI values ​​were calculated and compared with the measured PLI values ​​for verification. Table 6 summarizes the comparison between the verification results of the physiological index prediction model and the measured data.

[0131] Table 6. Validation results of the regression model for the recommended speed limit.

[0132] As shown in Table 6, the average measured PLI values ​​corresponding to the recommended speed limits range from 0.70 to 0.77, all within the moderate load range (0.60 to 0.80). Among them: Recommended medium speed conditions for a radius of 60~90m: 60m-35km / h: PLI average value = 0.76, close to the upper limit of the moderate load range, reflecting the high physiological pressure of the extreme radius; 70m-40km / h: PLI average value = 0.75, which is in the upper-middle position of the moderate load range; 80m-45km / h: Average PLI = 0.77, close to the upper limit of the moderate load range; 90m-50km / h: PLI average value = 0.73, which is in the middle of the moderate load range; The average PLI (Power Index) for these radii under high-speed conditions exceeds 0.82, indicating they have entered a high-load range: 0.86 for 60m-40km / h, 0.83 for 70m-45km / h, 0.87 for 80m-50km / h, and 0.82 for 90m-55km / h. Therefore, they are not suitable as recommended speed limits.

[0133] Recommended high-speed conditions (55km / h) within a 100m radius: 100m-55km / h: PLI average value = 0.70, which is in the lower-middle position of the moderate load range; The 100m radius is unique in that the favorable alignment conditions (radius reaching 2.5 times the design speed standard) significantly reduce the geometric constraints on cornering. Even at a relatively high speed of 55km / h, the average PLI remains at a moderate level of 0.70. In contrast, at a medium speed (50km / h) with a 100m radius, the average PLI is only 0.56, which has entered the lower limit of the light load range, indicating that road resources are not being fully utilized.

[0134] The relative deviation between the PLI value calculated by the physiological index prediction model and the measured mean PLI value ranged from -12.6% to +7.3%, with an average deviation of approximately 3.5%. Overall, the physiological index prediction model reflects the basic laws of PLI variation with radius and speed relatively well, and can provide theoretical support for the rationality of the speed limit value.

[0135] In summary, the average PLI of all recommended speed limit values ​​is controlled within the moderate load range of 0.70 to 0.77. This avoids both the distraction caused by excessively low load and the fatigue and error risks caused by excessively high load, thus verifying the scientific validity and consistency of the recommended speed limit scheme.

[0136] The core of this implementation method lies in first determining the recommended speed limit using actual measurement data, and then verifying it using a physiological index prediction model, with the two mutually reinforcing each other. The measured data directly reflects the driver's actual physiological response under different radius-speed combinations, avoiding potential biases in purely theoretical derivations. The physiological index prediction model mathematically quantifies the relationship between PLI (Physical Load Index) and radius and speed, avoiding the need for repeated driving simulations or real-vehicle tests for each new road segment, significantly reducing the time and economic cost of determining the speed limit. The average deviation between the model's calculated values ​​and the measured values ​​is only 3.5%, indicating that the physiological index prediction model can effectively predict overall physiological load.

[0137] The method for determining the speed limit value of small radius circular curves on mountain roads proposed in this invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of this invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for constructing a comprehensive physiological load index model for determining speed limit values, characterized in that, Includes the following steps: Step 1: Acquire multiple physiological indicators of multiple drivers as they pass through the target road section by using a multi-channel physiological signal acquisition device worn on the driver's body; Step 2: Process the collected multiple physiological indicators, calculate the reliability and individual variability of different physiological indicators based on the processed data, and statistically analyze the safety thresholds corresponding to different physiological indicators. Step 3: Based on the safety thresholds, reliability, and individual variability corresponding to different physiological indicators, construct a comprehensive physiological load index model.

2. The method for constructing a comprehensive physiological load index model for determining speed limit values ​​according to claim 1, characterized in that, The comprehensive physiological load index model is as follows: ； in, For heart rate growth rate, For skin cell growth rate, This is the root mean square value of electromyography. This is a comprehensive physiological load index.

3. A method for determining the speed limit value of a small-radius circular curve on a mountain highway, wherein the method is implemented based on the comprehensive physiological load index model construction method described in claim 1 or 2, characterized in that... Includes the following steps: Step A1: For different radii of circular curves on mountain roads, preset multiple vehicle speeds respectively; Step A2: Obtain multiple physiological indicators of the driver when passing through different circular curve radii on mountain roads at different vehicle speeds; Step A3: Use the multiple physiological indicators obtained in step A2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value for the radius of the corresponding circular curve of the mountain road based on the comprehensive physiological load index.

4. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 3, characterized in that, In step A1, multiple vehicle speeds are preset for different circular curve radii on mountain roads, specifically as follows: When the radius of the circular curve on the mountain road is 60m, the preset vehicle speeds are 30km / h, 35km / h and 40km / h. When the radius of the circular curve on the mountain road is 70m, the preset vehicle speeds are 35km / h, 40km / h and 45km / h. When the radius of the circular curve on the mountain road is 80m, the preset vehicle speeds are 40km / h, 45km / h and 50km / h. When the radius of the circular curve on the mountain road is 90m and 100m, the preset vehicle speed is 45km / h, 50km / h and 55km / h.

5. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 3, characterized in that, In step A3, the recommended speed limit value for the radius of the corresponding circular curve on the mountain road is determined based on the comprehensive physiological load index, specifically as follows: The comprehensive physiological load index is divided into multiple interval levels, and the vehicle operating speed that keeps the comprehensive physiological load index within the moderate load range is selected as the recommended speed limit value for the radius of the corresponding circular curve of the mountain road.

6. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 5, characterized in that, The comprehensive physiological load index is divided into multiple interval levels, specifically as follows: when When it is in the light load range, When the time is within the moderate load range, During the high-load period, when This is the overload zone.

7. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 4, characterized in that, The recommended speed limit is 35 km / h when the radius of a circular curve on a mountain road is 60 meters; 40 km / h when the radius is 70 meters; 45 km / h when the radius is 80 meters; 50 km / h when the radius is 90 meters; and 55 km / h when the radius is 100 meters.

8. A method for determining the speed limit value of a small-radius circular curve on a mountain highway, wherein the method is implemented based on the comprehensive physiological load index model construction method described in claim 1 or 2, characterized in that... Includes the following steps: Step B1: Obtain the radius of the circular curve of the mountain road and preset different vehicle speeds; Step B2: Construct a physiological index prediction model. Based on the radius of the circular curve of the mountain road and the preset different vehicle operating speeds, calculate the physiological parameters corresponding to the vehicle operating speed. Step B3: Use the physiological parameters calculated in step B2 as input to the comprehensive physiological load index model, calculate the comprehensive physiological load index, and determine the recommended speed limit value corresponding to the radius of the circular curve of the mountain road based on the comprehensive physiological load index.

9. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 8, characterized in that, The physiological indicator prediction models in step B2 include a heart rate growth rate prediction model, a skin conductance growth rate prediction model, and an electromyography root mean square value prediction model.

10. The method for determining the speed limit value of a small-radius circular curve on a mountain highway according to claim 9, characterized in that, The heart rate growth rate prediction model is as follows: ； The skin cell growth rate prediction model is as follows: ； The root mean square value prediction model for electromyography is as follows: ； in, The radius of the circular curve is... The vehicle's operating speed.

Images