Intelligent driving perception data evaluation system

Abstract

The invention, which relates to the technical field of driving perception data evaluation, discloses an intelligent driving perception data evaluation system comprising a wet and slippery bridge deck real-time monitoring module, a wet and slippery bridge deck perception information processing module, a road holding evaluation model construction module, a dynamic braking strategy adjustment module and a feedback optimization and model updating module. The wet and slippery bridge floor sensing information processing module is used for acquiring wet and slippery bridge floor sensing information in real time when a vehicle runs on the wet and slippery bridge surface at a high speed, analyzing after acquiring, and respectively generating a wet and slippery bridge floor friction coefficient and a road holding response lag index; and the road holding evaluation model building module is used for building a road holding evaluation model for the generated wet sliding bridge floor friction coefficient and the road holding response lag index, generating a road holding evaluation coefficient and analyzing the road holding evaluation coefficient after generation. The method solves the problem of braking lag caused by reduction of the road holding force on the wet and slippery bridge floor, and achieves the effects of evaluating the road holding force in real time and dynamically adjusting the braking strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of driving perception data evaluation, and in particular to an intelligent driving perception data evaluation system. Background Art

[0002] Driving perception refers to the process by which intelligent driving vehicles perceive the surrounding environment information in real time through a variety of sensors (such as cameras, lidar, radar, ultrasound, etc.), and identify and analyze key elements such as obstacles, pedestrians, traffic lights, lane lines, etc. In this process, the vehicle will generate a large amount of perception data, including the distance, speed, position, shape, category, lane line identification information of the target object, and the dynamic changes of the surrounding environment. These data directly affect the safety and decision-making ability of the vehicle, so it is crucial to evaluate the data generated by driving perception. Through evaluation, the accuracy, real-time and stability of vehicle perception can be quantified, possible perception errors and delays can be discovered, and the perception algorithm and sensor configuration can be optimized to ensure that intelligent driving vehicles have sufficient environmental adaptability and reaction speed in various complex scenarios, thereby improving driving safety and reliability.

[0003] Existing intelligent driving perception data evaluation technology is usually evaluated in detail through a multi-step comprehensive analysis. First, the vehicle's various sensors (such as cameras, lidar, radar, etc.) collect surrounding environment data in real time. These data include the location, speed, distance, lane lines and obstacles of the target. Then, the raw data is denoised, synchronized and formatted through the data preprocessing module to ensure the accuracy and consistency of the data. Next, the feature extraction module extracts key features, such as the vehicle's recognition accuracy of obstacles, response time, environmental adaptability, etc. After feature extraction, the system calculates multiple key parameters, such as perception error rate, recognition accuracy and response delay index. Finally, the evaluation module generates a perception capability evaluation coefficient by analyzing these parameters and compares it with the preset threshold to determine whether the vehicle's perception capability meets the requirements. The entire evaluation process not only focuses on the accuracy of the sensor, but also evaluates the stability and real-time performance of the perception system in various complex driving scenarios to ensure that the vehicle can drive safely in a changing environment.

[0004] The prior art has the following deficiencies:

[0005] When driving at high speed on a slippery bridge surface, the friction between the vehicle's tires and the road surface will drop significantly, and the tire grip will quickly weaken due to the formation of a water film on the road surface. In this case, the system needs to adjust the braking strategy in real time to ensure safety. However, due to the high speed of the vehicle, the friction coefficient between the slippery bridge surface and the ordinary road is quite different, and the system sensor cannot perceive the change in tire grip in time, resulting in perception lag. This lag makes it difficult for the system to quantify the impact of the slippery surface on the tire grip in real time, which in turn leads to insufficient braking response and difficulty in accurately controlling the braking distance. Ultimately, the system's misjudgment of grip may cause the vehicle to slide due to insufficient friction, increasing the risk of accidents or vehicle loss of control. Especially when driving at high speeds, inaccurate grip assessment will greatly increase the possibility of accidents.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent driving perception data evaluation system to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent driving perception data evaluation system, including a slippery bridge deck real-time monitoring module, a slippery bridge deck perception information processing module, a grip evaluation model building module, a dynamic braking strategy adjustment module, and a feedback optimization and model updating module;

[0009] The real-time monitoring module for slippery bridge decks monitors the slippery condition of the bridge deck and the driving speed of the vehicle in real time through a sensor network when driving at high speed on a slippery bridge surface. It also establishes a real-time monitoring framework for the slippery bridge deck grip assessment and braking response based on the change in friction force on the slippery bridge deck.

[0010] The slippery bridge surface perception information processing module acquires the slippery bridge surface perception information of the vehicle when it is driving at high speed on the slippery bridge surface in real time, and analyzes it after acquisition to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively;

[0011] A grip evaluation model building module builds a grip evaluation model based on the generated slippery bridge friction coefficient and grip response hysteresis index, generates a grip evaluation coefficient, and analyzes it after generation, and analyzes the current grip condition of the vehicle based on the generated grip evaluation coefficient, and divides it into sufficient grip, marginal grip, and insufficient grip;

[0012] Dynamic braking strategy adjustment module, which dynamically adjusts the braking strategy for different grip conditions based on the analysis results of the grip evaluation coefficient;

[0013] The feedback optimization and model update module continuously monitors the grip condition and braking response effect of the slippery bridge surface, adjusts the grip evaluation model and braking strategy in real time, and optimizes the braking control in slippery conditions through feedback data.

[0014] Preferably, the slippery bridge surface perception information processing module is used to obtain the slippery bridge surface perception information when the vehicle is driving at high speed on the slippery bridge surface in real time, and analyze it after acquisition to generate the slippery bridge surface friction coefficient and the grip response hysteresis index respectively. The specific analysis logic is as follows:

[0015] Real-time acquisition of slippery bridge surface perception information when a vehicle is driving at high speed on a slippery bridge surface, and pre-processing after acquisition;

[0016] Extracting slippery bridge surface friction information and grip response hysteresis information from the preprocessed slippery bridge surface perception information;

[0017] The extracted slippery bridge surface friction information and grip response hysteresis information are analyzed to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively.

[0018] Preferably, the logic for obtaining the slippery bridge surface friction coefficient is as follows:

[0019] The slippery bridge surface friction information is extracted from the preprocessed slippery bridge surface perception information, including the average water film thickness of the contact area between the vehicle tire and the bridge surface at different times over a period of time when the vehicle is driving at high speed on the slippery bridge surface, the actual speed of the vehicle, and the slip rate of the vehicle tire, and they are calibrated as HW m 、V m and S m , H.W. m V represents the average thickness of the water film in the contact area between the vehicle tire and the bridge surface at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the actual speed of the vehicle at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the slip rate of the vehicle tire at time m within a period of time when the vehicle is driving at high speed on a slippery bridge surface, m = 1, 2, 3, ..., g, g is a positive integer;

[0020] The empirical coefficients for correcting the influence of the average thickness of the water film in the contact area between the vehicle tire and the bridge surface, the actual speed of the vehicle and the slip rate of the vehicle tire on the friction force are obtained and calibrated as α, β and γ respectively;

[0021] Calculate the friction coefficient of the slippery bridge surface. The specific calculation formula is as follows:

[0022]

[0023] Where CFWBS is the friction coefficient of the wet bridge surface.

[0024] Preferably, the logic for obtaining the grip response hysteresis index is as follows:

[0025] The grip response hysteresis information in the preprocessed slippery bridge surface perception information is extracted, including the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit at different times within a period of time when the vehicle is driving at high speed on the slippery bridge surface, the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient, the time required for the central control unit to issue a braking command to the actual execution of the braking system, the average speed of the vehicle, and the total weight of the vehicle, and calibrated as ΔTS n , ΔTP n , ΔTB n , VP and ZG, ΔTS n It represents the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit within a period of time n when the vehicle is driving at high speed on a slippery bridge surface, ΔTP n It represents the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient at time n when the vehicle is driving at high speed on the slippery bridge surface. ΔTB n It represents the time required for the central control unit to issue a braking command to the actual execution of the braking system within a period of time n when the vehicle is driving at high speed on a slippery bridge surface. VP represents the average speed of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. ZG represents the total weight of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. n=1, 2, 3, …, k, where k is a positive integer.

[0026] Obtaining a correction coefficient for correcting the influence of the total weight of the vehicle and the average speed of the vehicle on the lag time, and calibrating it as δ after obtaining;

[0027] Calculate the grip response hysteresis index. The specific calculation formula is as follows:

[0028]

[0029] Where GRLI is the grip response hysteresis index.

[0030] Preferably, constructing a grip evaluation model based on the generated slippery bridge surface friction coefficient and grip response hysteresis index to generate a grip evaluation coefficient specifically includes the following steps:

[0031] Collect several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time and calibrate them as CFWBSx GRLI x and GEC x , x represents the serial numbers of several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time, x=3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time are formed into a historical data set;

[0032] The multivariate regression model is selected as the grip evaluation model, and trained through the historical data set to determine the value of the regression coefficient according to the formula:

[0033] GEC x =β0+β1*CFWBS x -β2*GRLI x

[0034] Where β0, β1 and β2 are regression coefficients;

[0035] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0036] The final determined regression coefficient is used to construct the grip evaluation model, and the real-time generated wet bridge friction coefficient CFWBS and grip response hysteresis index GRLI are input to generate the grip evaluation coefficient GEC in real time.

[0037] Preferably, the generated grip evaluation coefficient GEC is compared with a preset grip evaluation coefficient threshold interval [GEC min , GEC max ] and analyze the current grip condition of the vehicle based on the comparison results, and divide it into sufficient grip, marginal grip and insufficient grip. The specific comparison, analysis and division are as follows:

[0038] If GEC <GEC min , the current grip condition of the vehicle is insufficient grip;

[0039] If GEC min ≤GEC≤GEC max , the current grip condition of the vehicle is critical;

[0040] If GEC>GEC max , the vehicle's current grip condition is sufficient grip.

[0041] Preferably, in the dynamic braking strategy adjustment module, based on the analysis result of the grip evaluation coefficient, the braking strategy is dynamically adjusted according to different grip conditions, specifically:

[0042] When the grip condition is insufficient, the dynamic adjustment braking strategy is: take emergency braking measures immediately and send an emergency alarm to the driver to remind him that the current grip condition is insufficient;

[0043] When the grip condition is critical, the dynamic adjustment braking strategy is: take preventive braking strategy, remind the driver that the current grip condition is critical, and suggest driving with caution;

[0044] When the grip condition is sufficient, the dynamically adjusted braking strategy is: no braking strategy is required, and the vehicle continues to move in the current state.

[0045] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0046] 1. The technical solution in the present invention has demonstrated extremely high technical effects, especially in solving the problems of reduced grip and delayed braking response on slippery bridge surfaces. First, by real-time monitoring of the friction conditions of the slippery bridge surface and the operating status of the vehicle, the system can accurately capture the changes in friction between the tires and the road surface. By using the dynamic data of the sensor network, combined with the calculation of the slippery bridge surface friction coefficient and the grip response lag index, the system realizes an accurate assessment of grip at different time points. Such dynamic monitoring greatly improves the real-time perception capability of grip changes and avoids the misjudgment of grip caused by perception lag in traditional technologies.

[0047] 2. The present invention establishes a grip evaluation model. The system dynamically adjusts the grip evaluation coefficient according to different road conditions and vehicle speeds, and performs real-time analysis. The application of this multivariate regression model enables the system to flexibly respond to changes in grip under different environments and driving conditions, ensuring accurate distinction between sufficient, critical or insufficient grip when the vehicle is driving on a slippery bridge. Compared with traditional fixed evaluation models, this technology can more adaptively handle different combinations of bridge slipperiness and vehicle speeds, greatly improving the accuracy of grip evaluation.

[0048] 3. The biggest advantage of the present invention lies in its dynamic braking strategy adjustment and feedback optimization module. The system can not only make timely adjustments based on the current grip conditions, such as taking emergency braking, deceleration or reminding the driver, but also continuously optimize the braking strategy through the feedback mechanism. By collecting past braking data, the system can optimize the grip assessment model and adjust the strategy to improve the accuracy of the braking response. This adaptive feedback mechanism ensures the long-term stability and continuous optimization capabilities of the system, allowing the vehicle to always maintain optimal grip and braking control when driving on a slippery bridge, significantly reducing the risk of sliding out of control and improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0050] Figure 1 Schematic diagram of the modules of the intelligent driving perception data evaluation system of the present invention. DETAILED DESCRIPTION

[0051] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0052] The present invention provides Figure 1 The intelligent driving perception data evaluation system shown includes a slippery bridge surface real-time monitoring module, a slippery bridge surface perception information processing module, a grip evaluation model building module, a dynamic braking strategy adjustment module, and a feedback optimization and model update module;

[0053] The real-time monitoring module for slippery bridge decks monitors the slippery condition of the bridge deck and the driving speed of the vehicle in real time through a sensor network when driving at high speed on a slippery bridge surface. It also establishes a real-time monitoring framework for the slippery bridge deck grip assessment and braking response based on the change in friction force on the slippery bridge deck.

[0054] When driving at high speed on a slippery bridge surface, the system can monitor the slippery condition of the bridge deck in real time through sensors installed on the vehicle chassis and bridge surface. The specific implementation methods include: using humidity sensors and temperature sensors to measure the thickness of the water film and temperature changes on the bridge surface, and combining air pressure sensors to detect the humidity and rainfall in the air to determine whether the bridge deck is in a slippery state. At the same time, the vehicle speed is obtained in real time through the vehicle speed sensor. These data are integrated into the vehicle's central control system and transmitted to the central processing unit through the data communication protocol. Through the sensor network, the system can collect and synchronize this information at a high frequency, laying the foundation for subsequent grip evaluation. The data is preliminarily cleaned and filtered through the on-board processing system to ensure the real-time and accuracy of the monitoring information.

[0055] In order to solve the safety hazards caused by the change of grip on a slippery bridge deck, the system dynamically evaluates the friction change requirements of the slippery bridge deck by analyzing the slippery conditions of the bridge deck and the vehicle speed collected by the sensor. The specific implementation methods include: converting the slipperiness of the bridge deck into a friction change coefficient through a physical friction model, which calculates the actual friction between the tire and the road surface based on the slippery bridge deck data. The system combines the friction change data with the vehicle speed data, and generates the grip condition through a real-time grip evaluation algorithm. Then, the system establishes a braking response framework based on the evaluation, divides different grip scenarios (such as sufficient, critical, and insufficient), and triggers corresponding braking strategies according to different grip states. The entire process is executed through software algorithms, without the need for additional hardware, to achieve automatic adjustment and real-time response.

[0056] The role of the "wet bridge real-time monitoring module" is to solve the problem that the system cannot perceive the change of grip in time when driving at high speed on a wet bridge surface. On a wet bridge, due to the influence of factors such as water film, the friction between the tire and the road surface drops sharply, and the grip changes rapidly and complexly. It is difficult for conventional systems to evaluate this change in a timely and accurate manner, resulting in a lag in braking response and an increase in the risk of accidents. Therefore, by monitoring the slippery condition of the bridge surface and the speed of the vehicle in real time through the sensor network, the system can accurately perceive the current environment, dynamically generate the demand for friction changes, and establish a real-time grip evaluation and braking response framework. Through this software implementation, the system can continuously monitor the slippery road conditions and adjust the vehicle braking strategy according to environmental changes to avoid the brake lag caused by grip changes, thereby improving the safety of driving on wet bridges. This mechanism ensures that the vehicle can respond to insufficient grip in a timely manner when driving at high speeds, reduce the risk of loss of control, and meet the safety requirements of intelligent driving.

[0057] The slippery bridge surface perception information processing module acquires the slippery bridge surface perception information of the vehicle when it is driving at high speed on the slippery bridge surface in real time, and analyzes it after acquisition to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively;

[0058] In this embodiment, the slippery bridge surface perception information processing module is used to obtain the slippery bridge surface perception information of the vehicle when it is driving at high speed on the slippery bridge surface in real time, and analyze it after acquisition to generate the slippery bridge surface friction coefficient and the grip response hysteresis index respectively. The specific analysis logic is as follows:

[0059] Real-time acquisition of slippery bridge surface perception information when a vehicle is driving at high speed on a slippery bridge surface, and pre-processing after acquisition;

[0060] Real-time acquisition of slippery bridge surface perception information when the vehicle is driving at high speed on a slippery bridge surface is usually achieved through an on-board sensor network and remote data communication. The specific acquisition methods include: real-time acquisition of the slippery conditions of the bridge surface (such as water film thickness and temperature) through humidity sensors, temperature sensors and road surface condition sensors, and the vehicle speed and tire grip conditions through vehicle speed sensors and tire slip sensors. All these sensor data are transmitted to the central processing unit through the on-board network (such as CAN bus), which aggregates and synchronizes the data. In addition, the system can also interact with external infrastructure through wireless communication technology (such as 5G or V2X vehicle-road cooperative communication) to obtain more detailed road condition information to ensure accurate grip-related data when the bridge is slippery. The core of real-time data transmission is to ensure that data delay is minimized so that the vehicle can respond quickly to slippery conditions.

[0061] The main purpose of preprocessing is to ensure that the acquired perception information of slippery bridge deck is clean, accurate, and can adapt to the subsequent grip analysis and calculation. First, because sensors may be affected by noise, external interference or data acquisition errors in complex environments, the preprocessing steps usually include noise filtering (such as Kalman filtering algorithm, which is used to smooth signals and reduce sensor noise), data format standardization (converting data from different sources to the same data format and unit), and missing data interpolation (filling in missing data through historical data or extrapolation methods). The specific implementation of preprocessing is to process the input data stream in real time through software algorithms, eliminate abnormal data, and perform consistency correction on data from multiple sources to ensure that the perception information input to the system can accurately reflect the current slippery bridge deck state and vehicle grip. This process is executed by the automatic data processing module in the background to ensure that high-quality analysis data can be provided in the case of high-frequency sensor data input.

[0062] Extracting slippery bridge surface friction information and grip response hysteresis information from the preprocessed slippery bridge surface perception information;

[0063] Extracting the slippery bridge friction information and grip response hysteresis information from the preprocessed slippery bridge perception information can be achieved through feature extraction algorithms and data classification methods. First, in the preprocessed perception data, the system will automatically identify key variables related to friction, such as water film thickness, road roughness, tire slip rate, vehicle speed, etc., as part of the slippery bridge friction information based on the data source and the preset data type label through feature selection algorithms (such as PCA principal component analysis or feature engineering). Then, the system will extract the grip response hysteresis information by analyzing data related to the system reaction time, such as sensor delay, data processing time, and vehicle speed. The entire extraction process is achieved through logical judgment and feature clustering algorithms in the data flow pipeline, which decomposes the original data into different feature data sets, which are used for the calculation of friction and hysteresis response respectively. In this way, the system can accurately extract the required friction information and grip response hysteresis information according to specific rules, providing accurate basic data for subsequent calculations.

[0064] The extracted slippery bridge surface friction information and grip response hysteresis information are analyzed to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively.

[0065] In this embodiment, the logic for obtaining the friction coefficient of the slippery bridge surface is as follows:

[0066] The slippery bridge surface friction information is extracted from the preprocessed slippery bridge surface perception information, including the average water film thickness of the contact area between the vehicle tire and the bridge surface at different times over a period of time when the vehicle is driving at high speed on the slippery bridge surface, the actual speed of the vehicle, and the slip rate of the vehicle tire, and they are calibrated as HW m 、V m and S m , H.W. m V represents the average thickness of the water film in the contact area between the vehicle tire and the bridge surface at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the actual speed of the vehicle at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the slip rate of the vehicle tire at time m within a period of time when the vehicle is driving at high speed on a slippery bridge surface, m = 1, 2, 3, ..., g, g is a positive integer;

[0067] In the process of extracting friction information on slippery bridge decks, several types of key quantitative data can be obtained in real time through sensor networks and vehicle-mounted systems, and analyzed through software processing. First, the average thickness of the water film is obtained in real time through humidity sensors, rain sensors, and road surface detection devices installed under the vehicle chassis. These sensors can detect the thickness of the water film on the bridge surface during vehicle driving and transmit the data to the central control system for aggregation and average calculation. Secondly, the actual speed of the vehicle is obtained in real time through a vehicle speed sensor (such as a vehicle-mounted GPS or wheel speed sensor), which continuously records the speed information of the vehicle while driving and is updated and transmitted in real time through the vehicle control system. Finally, the tire slip rate is obtained in real time through a slip sensor or wheel speed difference sensor installed on the tire. The slip rate reflects the difference between tire slip and vehicle speed. This data can help determine whether the tire is slipping. All of these data are synchronously transmitted through the vehicle network (such as the CAN bus) and cleaned and processed by the real-time data processing module of the software to ensure that the friction information used for calculation is accurate and timely.

[0068] The empirical coefficients for correcting the influence of the average thickness of the water film in the contact area between the vehicle tire and the bridge surface, the actual speed of the vehicle and the slip rate of the vehicle tire on the friction force are obtained and calibrated as α, β and γ respectively;

[0069] The empirical coefficients used to correct the average thickness of the water film in the contact area between the vehicle tire and the bridge surface, the actual speed of the vehicle, and the slip rate of the vehicle tire on the friction are usually determined in advance through simulation of experimental data and historical scenarios, and do not need to be obtained dynamically in real time during the actual system operation. However, during the operation of the system, the calculation can be automatically adjusted based on these empirical coefficients set previously by the software system. Specifically, the α empirical coefficient is used to correct the effect of the water film thickness on the friction. The thicker the water film, the smaller the friction. This coefficient is pre-set by a large number of experiments based on the weakening effect of the water film on the friction. The β empirical coefficient is used to correct the effect of the vehicle speed on the friction. The faster the speed, the greater the degree of friction reduction due to the hydroplaning phenomenon. The γ empirical coefficient is used to correct the effect of the slip rate on the friction. The higher the slip rate, the more severe the tire slippage and the lower the friction. The purpose of obtaining these empirical coefficients is to correct the key variables that affect friction so that the friction model can accurately reflect the actual grip conditions of the vehicle in various environments, ensuring that the system can adjust the calculation results in real time according to the current water film thickness, vehicle speed and slip rate, thereby providing more accurate friction estimation and improving driving safety.

[0070] Calculate the friction coefficient of the slippery bridge surface. The specific calculation formula is as follows:

[0071]

[0072] Where CFWBS is the friction coefficient of the wet bridge surface.

[0073] The calculation formula of the slippery bridge friction coefficient CFWBS combines multiple key factors to calculate the actual friction under slippery bridge conditions in a weighted average manner. First, each term in the formula represents the slippery bridge friction coefficient at a different time point. By accumulating the friction at all times (m=1 to g), the average value (divided by g) is finally taken, which can reflect the dynamic changes of friction under slippery conditions over a period of time. Specifically, each part of the formula has a specific physical meaning: 1-α*HW m ) is used to correct the water film thickness (HW m ) The effect on friction: the thicker the water film, the smaller the friction; Used to correct the vehicle speed V m The influence of friction force, the faster the vehicle speed, the more serious the hydroplaning phenomenon, and the more obvious the decrease in friction force; finally, (1-γ*S m ) is used to correct the tire slip rate (S m ) on friction. The greater the slip rate, the smaller the friction. After these factors are multiplied, they are accumulated and averaged moment by moment to ensure that the system can dynamically reflect the changes in friction at different time points, and finally generate a comprehensive wet bridge friction coefficient. This calculation method can effectively capture changes in the environment and vehicle status, ensure the accuracy of friction estimation, and thus help the system better evaluate grip and adjust braking strategies.

[0074] The relationship between the friction coefficient CFWBS of a slippery bridge surface and the analysis of the current grip condition of the vehicle is that the friction coefficient is the core parameter that reflects the grip strength between the tire and the bridge surface. When the friction coefficient of a slippery bridge surface is large, it means that the friction between the tire and the road surface is strong, the vehicle has sufficient grip, and can drive safely; conversely, when the friction coefficient is small, it indicates that the friction between the tire and the bridge surface is weakened, the vehicle's grip is insufficient, and it is easy to slip or lose control. By analyzing the friction coefficient of a slippery bridge surface, the system can evaluate the current grip condition of the vehicle in real time, and dynamically adjust the braking strategy based on the evaluation results to ensure the vehicle's driving safety in slippery environments. Therefore, the friction coefficient directly determines the adequacy of the vehicle's current grip and is an important basis for the intelligent driving system to make decisions.

[0075] In this embodiment, the logic for obtaining the grip response hysteresis index is as follows:

[0076] The grip response hysteresis information in the preprocessed slippery bridge surface perception information is extracted, including the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit at different times within a period of time when the vehicle is driving at high speed on the slippery bridge surface, the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient, the time required for the central control unit to issue a braking command to the actual execution of the braking system, the average speed of the vehicle, and the total weight of the vehicle, and calibrated as ΔTS n , ΔTP n , ΔTB n , VP and ZG, ΔTS n It represents the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit within a period of time n when the vehicle is driving at high speed on a slippery bridge surface, ΔTP n It represents the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient at time n when the vehicle is driving at high speed on the slippery bridge surface. ΔTB n It represents the time required for the central control unit to issue a braking command to the actual execution of the braking system within a period of time n when the vehicle is driving at high speed on a slippery bridge surface. VP represents the average speed of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. ZG represents the total weight of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. n=1, 2, 3, …, k, where k is a positive integer.

[0077] When extracting the grip response hysteresis information from the slippery bridge surface perception information, several types of key quantitative data can be obtained in real time through the vehicle sensor network and the central processing system, and processed by the software system. First, the sensor data transmission delay time (the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit) can be monitored in real time through the data transmission link delay between the vehicle sensor and the central control unit. When the sensor on the vehicle collects the slippery bridge surface information, it transmits the data to the central control system through the vehicle bus (such as CAN bus) or wireless communication network. The software estimates the delay time by measuring the round-trip time of the data transmission. Secondly, the central control unit data processing time (the time spent by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient) can be calculated in real time through the internal timing mechanism of the software. After receiving the sensor data, the system performs the analysis of the friction coefficient and other processing time is determined by the processor performance and task complexity of the control unit. The software can obtain the data processing time by recording the timestamps at the start and end of the analysis. The braking system response time (the time required from the central control unit issuing a braking command to the actual execution of the braking system) is obtained by measuring the time difference between the central control system sending a braking command and the braking system executing it. The system can detect the execution status of the braking command in real time through an internal feedback mechanism. The average speed of the vehicle is obtained in real time through a speed sensor (such as a wheel speed sensor or GPS), and a relatively stable speed value is calculated from the average speed of the vehicle while it is traveling. The gross vehicle weight is usually a preset value for the vehicle, which is input from the vehicle's factory data or load monitoring system when the system is started and used as a constant. All of these data are automatically collected and processed by the software, which can provide accurate input for calculating the grip response lag index, thereby evaluating the response speed and lag time of the system.

[0078] Obtaining a correction coefficient for correcting the influence of the total weight of the vehicle and the average speed of the vehicle on the lag time, and calibrating it as δ after obtaining;

[0079] The correction coefficients used to correct the effects of the gross vehicle weight and average vehicle speed on the lag time are usually determined in advance through historical experimental data or simulation analysis, and are preset as constants in the software system, rather than dynamic values ​​obtained in real time. However, during system operation, the software automatically applies these empirical coefficients to correct the lag time based on the gross vehicle weight and average speed. Specifically, the correction coefficient for the gross vehicle weight is used to reflect the effect of the vehicle load on the system response lag. The heavier the vehicle, the greater the inertia, and the longer the lag time may be, so this coefficient adjusts the effect of weight on the response time. The correction coefficient for the average vehicle speed is used to correct the effect of speed on the lag time. The faster the speed, the longer the vehicle's reaction time may be when braking, because the grip decreases at high speeds and the braking time increases. The purpose of obtaining these empirical coefficients is to ensure that the system can automatically adjust the calculation of the lag time according to the different weights and speeds of the vehicle, so as to more accurately reflect the actual response of the system and ensure the effectiveness and safety of the braking operation. Through the automatic correction mechanism in the software, these coefficients can be dynamically applied to the calculation of the lag index to ensure that the system can perform accurate lag time evaluation under a variety of different working conditions.

[0080] Calculate the grip response hysteresis index. The specific calculation formula is as follows:

[0081]

[0082] Where GRLI is the grip response hysteresis index.

[0083] The calculation formula of the grip response hysteresis index GRLI combines multiple variables to accurately evaluate the braking response hysteresis of a vehicle when driving at high speed on a slippery bridge. First, ΔTS in the formula n , ΔTP n and ΔTB n They represent the sensor data delay, data processing time, and braking system response time at different time points respectively. The accumulation of these three reflects the total response lag of the system at each moment. By accumulating n = 1 to k time points and taking the average (divided by k), the average lag value over a period of time can be obtained. Secondly, the average vehicle speed VP and the gross vehicle weight ZG affect the inertia and braking effect of the vehicle on a slippery bridge deck, respectively. Therefore, the formula combines the weight and speed of the vehicle to correct the lag time through the empirical correction coefficient δ. The higher the speed and the heavier the load, the longer the system lag time may be. Therefore, VP is placed in the denominator for adjustment. At the same time, Used to correct the effect of weight on hysteresis. This calculation method ensures that the hysteresis index can reflect the combined effects of different vehicle speeds, loads and system response times, thus providing a scientific basis for the dynamic adjustment of grip and braking response.

[0084] The relationship between the grip response hysteresis index GRLI and the analysis of the vehicle's current grip condition is that GRLI is used to quantify the time delay from the system's perception of grip changes to actual braking operations, which directly affects the vehicle's grip performance in slippery conditions. The larger the hysteresis index, the longer the system takes to respond to grip changes, the lower the utilization rate of grip, and the vehicle may not be able to adjust the braking operation in time on slippery roads, resulting in insufficient grip and increasing the risk of the vehicle skidding or losing control. By analyzing the grip response hysteresis index, the system can evaluate the braking effectiveness of the vehicle under the current grip condition, so as to take necessary preventive measures, such as adjusting the driving speed or triggering the braking command in advance, to ensure that the vehicle can drive safely in slippery environments. Therefore, GRLI directly affects the effective use of grip and is a key evaluation indicator to ensure the safety of vehicles under complex road conditions.

[0085] A grip evaluation model building module builds a grip evaluation model based on the generated slippery bridge friction coefficient and grip response hysteresis index, generates a grip evaluation coefficient, and analyzes it after generation, and analyzes the current grip condition of the vehicle based on the generated grip evaluation coefficient, and divides it into sufficient grip, marginal grip, and insufficient grip;

[0086] In this embodiment, a grip evaluation model is constructed based on the generated slippery bridge surface friction coefficient and grip response hysteresis index to generate a grip evaluation coefficient, which specifically includes the following steps:

[0087] Collect several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time and calibrate them as CFWBS x GRLI x and GEC x , x represents the serial numbers of several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time, x=3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time are formed into a historical data set;

[0088] The collection of the slippery bridge friction coefficient, grip response hysteresis index and corresponding grip evaluation coefficient generated in the past period of time can be achieved through the data recording module in the software system. The specific method is: when the system calculates the slippery bridge friction coefficient and grip response hysteresis index in real time, the calculation results at each moment are automatically recorded in the data storage module. The system stores these data according to timestamps to ensure that they can be accurately traced back to the specific values ​​within a specific time period. The software can save these values ​​regularly through databases or distributed storage technologies, and store the slippery bridge friction coefficient and grip response hysteresis index together with the grip evaluation coefficient actually calculated, to ensure that these data can be retrieved in subsequent analysis and model training. Through such automated data storage and collection methods, the system can continuously and reliably accumulate historical data, providing sufficient reference for the subsequent training and optimization of the grip evaluation model.

[0089] The value of x is a positive integer greater than or equal to 3 because there are three variables (friction coefficient of slippery bridge surface, grip response hysteresis index, grip evaluation coefficient), and three regression coefficients β0, β1 and β2 need to be solved, so at least three independent equations are needed to accurately calculate these three regression coefficients. Based on this, the value of x must be greater than or equal to 3 to ensure that we have enough equations for regression analysis. To explain further, each equation in the regression analysis corresponds to a data point (a combination of slippery bridge surface friction coefficient, grip response hysteresis index and grip evaluation coefficient). The more these data points, the better the model fits, and the better it can capture the changing trend of grip evaluation under different conditions. Therefore, the value of x is greater than or equal to 3, not only to meet the mathematical solution requirements, but also to ensure that there are enough data to improve the accuracy and robustness of the model and ensure the rationality and robustness of the regression coefficients.

[0090] The multivariate regression model is selected as the grip evaluation model, and trained through the historical data set to determine the value of the regression coefficient according to the formula:

[0091] GEC x =β0+β1*CFWBS x -β2*GRLI x

[0092] Where β0, β1 and β2 are regression coefficients;

[0093] The multivariate regression model is chosen as the grip evaluation model because grip evaluation involves the influence of multiple independent variables (wet bridge friction coefficient and grip response hysteresis index) on a dependent variable (grip evaluation coefficient), which is the strength of the multivariate regression model. The multivariate regression model is a statistical method used to analyze the joint influence of multiple independent variables on the dependent variable. It can train the model through historical data, identify the degree of influence of each variable on the target result under different conditions, and adjust it through the regression coefficient. In the formula, β0 is a constant term, which represents the basic value of the grip evaluation coefficient when the wet bridge friction coefficient and the grip response hysteresis index are zero; β1 is the regression coefficient of the wet bridge friction coefficient, which represents the contribution of this coefficient to grip evaluation; β2 is the regression coefficient of the grip response hysteresis index, which represents the negative impact of lag time on grip evaluation. Through the multivariate regression model, we can calculate these three regression coefficients, so that the model can accurately evaluate the current grip condition of the vehicle under different conditions, thereby optimizing driving safety.

[0094] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0095] The purpose of optimizing the regression coefficient by minimizing the error between the predicted value and the actual value is to ensure that the prediction accuracy of the model can reach the best level. The purpose of this is to find the most appropriate regression coefficients β0, β1 and β2 through historical data training, so that the model can more accurately estimate the grip evaluation coefficient in future predictions. This process is usually achieved through optimization algorithms such as the least squares method in the software. This method calculates the difference between the predicted value and the actual value through repeated iterations, and adjusts the regression coefficient to minimize the error. The specific implementation method is that during the model training process, the software automatically calculates the grip evaluation coefficient for each prediction, and then compares it with the actual grip value in the historical real data set to calculate the error, and adjusts the value of the regression coefficient through the algorithm until the error converges to the minimum value. This process is automatically completed through the iterative optimization module in the software to ensure that the model can accurately predict the grip condition of the vehicle after training.

[0096] The final determined regression coefficient is used to construct the grip evaluation model, and the real-time generated wet bridge friction coefficient CFWBS and grip response hysteresis index GRLI are input to generate the grip evaluation coefficient GEC in real time.

[0097] In this embodiment, the generated grip evaluation coefficient GEC is compared with the preset grip evaluation coefficient threshold interval [GEC min , GEC max] and analyze the current grip condition of the vehicle based on the comparison results, and divide it into sufficient grip, marginal grip and insufficient grip. The specific comparison, analysis and division are as follows:

[0098] If GEC <GEC min , the current grip condition of the vehicle is insufficient grip;

[0099] When GEC <GEC min , which means that the friction between the vehicle and the road is too low and the grip is seriously insufficient. At this time, the tires cannot effectively maintain sufficient contact with the road, which may cause the vehicle to slip or lose control when braking or turning. Especially in slippery, icy or other bad road conditions, insufficient grip will significantly increase the possibility of traffic accidents. When the system detects this situation, it must immediately take emergency measures, such as deceleration, warning or automatic emergency braking, to try to avoid accidents.

[0100] If GEC min ≤GEC≤GEC max , the current grip condition of the vehicle is critical;

[0101] When GEC min ≤GEC≤GEC max , indicating that the vehicle's current grip is in a critical state, meaning that the grip between the vehicle and the road is acceptable, but has approached the safety critical point. At this time, although the vehicle can still maintain control, once an emergency (such as sudden braking or sudden turning) is encountered, the grip may not be sufficient to cope with it. In this case, the system should issue a warning and advise the driver to drive carefully, slow down and avoid sudden driving actions to ensure that the vehicle stability does not further decrease.

[0102] If GEC>GEC max , the vehicle's current grip condition is sufficient grip.

[0103] When GEC>GEC max , indicating that the vehicle currently has sufficient grip, the tires maintain good friction with the road, and the vehicle can drive stably and safely. At this point, the system assesses that the vehicle's braking, acceleration, and steering are all within the controllable range, and driving safety is high. In this state, the vehicle can continue to drive at normal speed, and the system does not need to take special measures, but only needs to routinely monitor the grip condition to respond to any potential changes in a timely manner.

[0104] The pre-set grip evaluation coefficient threshold range can be determined by analyzing a large amount of historical driving data, experimental data and simulation data, and adjusted and optimized through software. The specific method is: the system first collects relevant data from the grip evaluation results under various different environments (such as wet, dry, ice and snow, etc.), and combines the vehicle's grip evaluation coefficient with the actual driving safety situation for correlation analysis. Through the statistical analysis tools in the software, such as regression analysis or classification algorithms, the system can find out the critical value ranges of insufficient, critical and sufficient grip under different conditions. Then, these results will be repeatedly optimized to make GEC min and GEC max This process meets the safety requirements of actual driving. Through model training and automatic adjustment, this process ensures that the set threshold is applicable to different vehicle models, road conditions and environments, and ultimately generates a scientific and reasonable grip evaluation coefficient range to ensure that the system can accurately judge the vehicle's grip status.

[0105] Dynamic braking strategy adjustment module, which dynamically adjusts the braking strategy for different grip conditions based on the analysis results of the grip evaluation coefficient;

[0106] In this embodiment, in the dynamic braking strategy adjustment module, based on the analysis result of the grip evaluation coefficient, the braking strategy is dynamically adjusted according to different grip conditions, specifically:

[0107] When the grip condition is insufficient, the dynamic adjustment braking strategy is: take emergency braking measures immediately and send an emergency alarm to the driver to remind him that the current grip condition is insufficient;

[0108] In the case of insufficient grip, the system can detect the current grip condition between the vehicle and the road in real time through the on-board sensor network (such as tire slip sensors, accelerometers, etc.), and initiate the emergency braking strategy through the emergency braking module in the software. The specific implementation method is: once the system calculates that the grip evaluation coefficient is lower than the threshold, the software automatically activates the emergency braking function, triggers the anti-lock braking system (ABS) and the vehicle stability control system (ESC), and transmits the braking signal to the brake system for rapid response. In addition, the system will issue an emergency warning to the driver through the in-car display or alarm sound, indicating insufficient grip and requiring the driver to take immediate measures. The purpose of this is to minimize the risk of sliding and loss of control of the vehicle in the case of insufficient grip, and ensure that the vehicle stops or slows down safely.

[0109] When the grip condition is critical, the dynamic adjustment braking strategy is: take preventive braking strategy, remind the driver that the current grip condition is critical, and suggest driving cautiously and avoid sudden acceleration or steering operations;

[0110] When the grip is in a critical state, the system can prevent risks through the preventive braking control module in the software. The specific implementation method is: when the grip evaluation coefficient is in the critical range, the system will control the speed of the vehicle through automatic deceleration or acceleration limit function to reduce the risk of possible skidding. Through the speed control algorithm in the software, the system can monitor the vehicle speed in real time and ensure that the vehicle is within a safe speed range by adjusting the throttle and brake force. At the same time, the system reminds the driver through the in-car display that the grip is in a critical state and advises the driver to avoid sudden turns or sudden acceleration. The purpose of this preventive strategy is to ensure that the vehicle is still controllable in potentially dangerous situations and prevent further deterioration of grip.

[0111] When the grip condition is sufficient, the dynamically adjusted braking strategy is: no braking strategy is required, and the vehicle continues to move in the current state.

[0112] When there is sufficient grip, the system does not need to take special braking measures, but it still needs to maintain continuous monitoring of grip. The specific implementation method is: the software continuously monitors the friction coefficient of the slippery bridge surface and the grip response hysteresis index through the on-board sensors, and calculates the grip evaluation coefficient in real time. The system will maintain real-time data collection and analysis of the vehicle and the environment through the normal monitoring module running in the background to ensure that the grip condition is always sufficient. In this case, the vehicle can travel at normal speed, but the system will still retain the real-time monitoring function so that corresponding measures can be taken immediately when the grip condition changes. The reason for doing this is to ensure that the vehicle is still under safe monitoring when there is sufficient grip, and can quickly respond to any sudden changes to ensure driving safety.

[0113] The feedback optimization and model update module continuously monitors the grip condition and braking response effect of the slippery bridge surface, adjusts the grip evaluation model and braking strategy in real time, and optimizes the braking control in slippery conditions through feedback data.

[0114] The software system collects grip data and the vehicle's braking response effect on slippery bridge surfaces in real time through the on-board sensor network. The grip condition is quantified by the slippery bridge friction coefficient and the grip response hysteresis index. The specific implementation method is that the software system regularly obtains data from the sensor, updates the vehicle's current grip condition in real time, and monitors the response of each braking, including braking distance, vehicle slip rate, and deceleration rate. The system compares this data with the expected response effect through the data analysis module that runs continuously in the background to ensure that the grip and braking response are always within a safe range. If the system detects that the grip condition is lower than expected or the braking response is not ideal, the feedback optimization mechanism is immediately triggered. The purpose of this continuous monitoring is to ensure that the system can respond quickly to changes in grip under complex road conditions, so as to make adjustments in advance and avoid dangerous situations.

[0115] When the system detects that the grip evaluation model does not match the actual road conditions, the software will automatically adjust the model. This is achieved by comparing the actual results of each braking with the predicted values ​​through feedback data, calculating the error, and automatically adjusting the parameters in the model, such as the weight of the slippery bridge friction coefficient and the response hysteresis index, or correcting certain empirical coefficients. The adjusted model can more accurately reflect the grip conditions of the current road surface and provide a more suitable braking strategy. For example, when a vehicle is driving on a slippery bridge, if it is found that the friction has dropped significantly, the system will improve the sensitivity of the braking response through an optimization algorithm. The purpose of this is to enable the grip evaluation model to always maintain a high degree of adaptability to the current environment, ensure that the braking strategy can be automatically optimized according to changes in road conditions, and reduce the risk of accidents.

[0116] The software system continuously optimizes the braking control strategy through a feedback mechanism to enhance the system's intelligent decision-making capabilities. Specifically, the system compares the actual data after each braking (such as braking distance, tire slip rate, etc.) with the model prediction value, calculates the error, and gradually reduces this error through machine learning or feedback-based control algorithms. For example, if it is found that the braking distance is longer than expected, the system will automatically adjust the braking intensity, speed limit or braking trigger conditions, and gradually optimize the control parameters. In addition, the system can further adjust the overall braking model through long-term accumulated data, so that its grip response under different road conditions is more accurate and stable. The purpose of this feedback mechanism is to ensure that the system can always maintain the best braking response effect through continuous self-optimization, especially in high-risk environments such as slippery roads, to improve driving safety.

[0117] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0118] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0119] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0120] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0121] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0124] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Intelligent driving perception data evaluation system, characterized by: It includes a real-time monitoring module for slippery bridge decks, a processing module for slippery bridge deck perception information, a grip evaluation model building module, a dynamic braking strategy adjustment module, and a feedback optimization and model update module; The real-time monitoring module for slippery bridge decks monitors the slippery condition of the bridge deck and the driving speed of the vehicle in real time through a sensor network when driving at high speed on a slippery bridge surface. It also establishes a real-time monitoring framework for the slippery bridge deck grip assessment and braking response based on the change in friction force on the slippery bridge deck. The slippery bridge surface perception information processing module acquires the slippery bridge surface perception information of the vehicle when it is driving at high speed on the slippery bridge surface in real time, and analyzes it after acquisition to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively; A grip evaluation model building module builds a grip evaluation model based on the generated slippery bridge friction coefficient and grip response hysteresis index, generates a grip evaluation coefficient, and analyzes it after generation, and analyzes the current grip condition of the vehicle based on the generated grip evaluation coefficient, and divides it into sufficient grip, marginal grip, and insufficient grip; Dynamic braking strategy adjustment module, based on the analysis results of grip evaluation coefficient, dynamically adjusts the braking strategy for different grip conditions; The feedback optimization and model update module continuously monitors the grip condition and braking response effect of the slippery bridge surface, adjusts the grip evaluation model and braking strategy in real time, and optimizes the braking control in slippery conditions through feedback data.

2. The intelligent driving perception data evaluation system according to claim 1, characterized in that: The slippery bridge surface perception information processing module is used to obtain the slippery bridge surface perception information of the vehicle when it is driving at high speed on the slippery bridge surface in real time, and analyze it after acquisition to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively. The specific analysis logic is as follows: Real-time acquisition of slippery bridge surface perception information when a vehicle is driving at high speed on a slippery bridge surface, and pre-processing after acquisition; Extracting slippery bridge surface friction information and grip response hysteresis information from the preprocessed slippery bridge surface perception information; The extracted slippery bridge surface friction information and grip response hysteresis information are analyzed to generate the slippery bridge surface friction coefficient and grip response hysteresis index respectively.

3. The intelligent driving perception data evaluation system according to claim 2, characterized in that: The logic for obtaining the slippery bridge surface friction coefficient is as follows: The slippery bridge surface friction information is extracted from the preprocessed slippery bridge surface perception information, including the average water film thickness of the contact area between the vehicle tire and the bridge surface at different times over a period of time when the vehicle is driving at high speed on the slippery bridge surface, the actual speed of the vehicle, and the slip rate of the vehicle tire, and they are calibrated as HW m 、V m and S m , H.W. m V represents the average thickness of the water film in the contact area between the vehicle tire and the bridge surface at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the actual speed of the vehicle at time m when the vehicle is driving at high speed on a slippery bridge surface. m It represents the slip rate of the vehicle tire at time m within a period of time when the vehicle is driving at high speed on a slippery bridge surface, m = 1, 2, 3, ..., g, g is a positive integer; The empirical coefficients for correcting the influence of the average thickness of the water film in the contact area between the vehicle tire and the bridge surface, the actual speed of the vehicle and the slip rate of the vehicle tire on the friction force are obtained and calibrated as α, β and γ respectively; Calculate the friction coefficient of the slippery bridge surface. The specific calculation formula is as follows: Where CFWBS is the friction coefficient of the wet bridge surface.

4. The intelligent driving perception data evaluation system according to claim 3, characterized in that: The acquisition logic of the grip response hysteresis index is as follows: The grip response hysteresis information in the preprocessed slippery bridge surface perception information is extracted, including the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit at different times within a period of time when the vehicle is driving at high speed on the slippery bridge surface, the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient, the time required for the central control unit to issue a braking command to the actual execution of the braking system, the average speed of the vehicle, and the total weight of the vehicle, and calibrated as ΔTS n , ΔTP n , ΔTB n , VP and ZG, ΔTS n It represents the delay time required for the sensor equipment on the vehicle to transmit the real-time collected information to the central processing unit within a period of time n when the vehicle is driving at high speed on a slippery bridge surface, ΔTP n It represents the time taken by the central control unit to analyze the slippery bridge surface perception information and generate the friction coefficient at time n when the vehicle is driving at high speed on the slippery bridge surface. ΔTB n It represents the time required for the central control unit to issue a braking command to the actual execution of the braking system within a period of time n when the vehicle is driving at high speed on a slippery bridge surface. VP represents the average speed of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. ZG represents the total weight of the vehicle when the vehicle is driving at high speed on a slippery bridge surface. n=1, 2, 3, …, k, where k is a positive integer. Obtaining a correction coefficient for correcting the influence of the total weight of the vehicle and the average speed of the vehicle on the lag time, and calibrating it as δ after obtaining; Calculate the grip response hysteresis index. The specific calculation formula is as follows: Where GRLI is the grip response hysteresis index.

5. The intelligent driving perception data evaluation system according to claim 4, characterized in that: A grip evaluation model is constructed based on the generated slippery bridge surface friction coefficient and grip response hysteresis index to generate a grip evaluation coefficient, which specifically includes the following steps: Collect several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time and calibrate them as CFWBS x GRLI x and GEC x , x represents the serial numbers of several slippery bridge friction coefficients, grip response hysteresis indexes and corresponding grip evaluation coefficients generated in the past period of time, x=3, 4, 5, ..., d, d is a positive integer, and the data collected in the past period of time are formed into a historical data set; The multivariate regression model is selected as the grip evaluation model, and trained through the historical data set to determine the value of the regression coefficient according to the formula: GEC x =β0+β1*CFWBS x -β2*GRLI x Where β0, β1 and β2 are regression coefficients; By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined; The final determined regression coefficient is used to construct the grip evaluation model, and the real-time generated wet bridge friction coefficient CFWBS and grip response hysteresis index GRLI are input to generate the grip evaluation coefficient GEC in real time.

6. The intelligent driving perception data evaluation system according to claim 5, characterized in that: The generated grip evaluation coefficient GEC is compared with the pre-set grip evaluation coefficient threshold range [GEC min , GEC max ] and analyze the current grip condition of the vehicle based on the comparison results, and divide it into sufficient grip, marginal grip and insufficient grip. The specific comparison, analysis and division are as follows: If GEC <GEC min , the current grip condition of the vehicle is insufficient grip; If GEC min ≤GEC≤GEC max , the current grip condition of the vehicle is critical; If GEC>GEC max , the vehicle’s current grip condition is sufficient grip.

7. The intelligent driving perception data evaluation system according to claim 6, characterized in that: In the dynamic braking strategy adjustment module, based on the analysis results of the grip evaluation coefficient, the braking strategy is dynamically adjusted according to different grip conditions, specifically: When the grip condition is insufficient, the dynamic adjustment braking strategy is: take emergency braking measures immediately and send an emergency alarm to the driver to remind him that the current grip condition is insufficient; When the grip condition is critical, the dynamic adjustment braking strategy is: take preventive braking strategy, remind the driver that the current grip condition is critical, and suggest driving with caution; When the grip condition is sufficient, the dynamically adjusted braking strategy is: no braking strategy is required, and the vehicle continues to move in the current state.

Images