A method, system, device and storage medium for safety warning of driving on a curved road

Abstract

The application relates to the technical field of driving safety early warning, and particularly discloses a curve driving safety early warning method, a curve driving safety early warning system, a curve driving safety early warning device and a storage medium. The early warning method comprises the following steps: acquiring real-time driving state information of a vehicle, road parameter information of a curve where the vehicle is located and environment state information; calculating a first critical speed at which the vehicle does not slide on the curve and a second critical speed at which the vehicle does not roll over according to the road parameter information of the curve; acquiring real-time load information of the vehicle, and determining a wind speed correction coefficient and a load correction coefficient according to the wind speed and the real-time load information; calculating a safe vehicle speed of the vehicle in the curve; and determining an early warning strategy and executing the early warning strategy according to the safe vehicle speed, the real-time driving state information of the vehicle, the environment state information and driver state information. The application can deeply integrate driver characteristics, real-time vehicle states and multi-dimensional environment information, and can dynamically learn and optimize based on historical big data.

Description

Technical Field

[0001] This invention relates to the field of driving safety warning technology, and in particular to a method, system, device and storage medium for driving safety warning on curves. Background Technology

[0002] With the rapid development of intelligent transportation systems, vehicle-road cooperative technology provides a new solution for improving road traffic safety, especially driving safety under complex road conditions, through real-time information interaction between vehicles, roads, and the cloud. Curves are high-incidence areas for traffic accidents. Due to blind spots, centrifugal force, and other factors, vehicles are prone to skidding, overturning, or colliding with oncoming or same-direction obstacles when driving on curves.

[0003] Existing curve safety warning technologies primarily rely on roadside sensors (such as radar and cameras) and onboard units to provide drivers with visual or auditory warnings by calculating safe speeds or safe distances; however, current technologies generally have the following limitations: I. Fixed Early Warning Models, Lacking Personalization and Adaptability: Existing safe distance or safe speed calculation models mostly use fixed parameters. For example, driver reaction time in safe distance calculation is usually a statistical average or only roughly divided into a few types, failing to reflect the dynamic differences between different drivers (such as aggressive and cautious drivers) or the same driver in different states in real time. At the same time, the models do not adequately consider environmental factors (such as crosswinds, visibility, and road surface micro-wear), and usually only use a limited adhesion coefficient table for static correction, making it difficult to adapt to instantaneous changes in weather and road conditions. This results in early warning thresholds that are either too conservative (leading to frequent false alarms) or not sensitive enough (increasing the risk of missed alarms).

[0004] Second, the perception dimension is singular and the decision factors are not comprehensive: Most solutions focus on vehicle status and road geometry parameters, and lack refined real-time perception and integration of key environmental factors that affect vehicle dynamic performance. For example, the impact of crosswinds on the stability of vehicles when cornering at high speeds and the impact of changes in actual vehicle load on the center of gravity and rollover threshold are rarely included in the calculation loop of safe speed. This means that the calculated safe speed may not be safe in certain harsh environments.

[0005] Third, the level of intelligence in early warning and control needs to be improved: existing systems mostly use early warning as the endpoint, and the control strategies are relatively simple and direct. They usually use a fixed overspeed ratio (such as 1.2 times the safe speed) to trigger a fixed level of alarm or braking, lacking risk-based refined and forward-looking decision-making; the system fails to make full use of the wide-area historical data (such as accident statistics of specific curves and common risk types) provided by the vehicle-road cooperative network to dynamically optimize the local early warning strategy, and the intelligence and learning evolution capabilities are insufficient.

[0006] One or more of the aforementioned technical defects urgently need to be addressed. Summary of the Invention

[0007] In view of this, the present invention aims to solve one or more of the above-mentioned technical defects and provide a curve driving safety warning system that can deeply integrate driver characteristics, real-time vehicle status, and multi-dimensional environmental information, and can dynamically learn and optimize based on historical big data.

[0008] In a first aspect, the present invention provides a method for warning of driving safety on curves, applied to an in-vehicle terminal, the method comprising: The system acquires real-time driving status information of the vehicle, road parameter information of the curve in which the vehicle is located, and environmental status information; the environmental status information includes at least one or more of the following: wind speed, visibility, duration of rainfall, and road surface wear. Based on the road parameter information of the curve, calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over in the curve. Obtain the real-time load information of the vehicle, and determine the wind speed correction coefficient and load correction coefficient based on the wind speed and the real-time load information, respectively. The safe speed of the vehicle in the curve is calculated based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient. Based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information, a warning strategy is determined and executed.

[0009] Furthermore, the step of calculating the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient includes: The safe vehicle speed is calculated using the following formula. : ; in, This is the first critical velocity. This is the second critical velocity; The wind speed correction factor is calculated using the following formula: ; in For real-time wind speed, The reference wind speed; The load correction factor is calculated using the following formula: ; Where L represents the real-time load of the vehicle. This is the rated load.

[0010] Furthermore, determining the warning strategy based on the safe vehicle speed, the vehicle's real-time driving status information, the environmental status information, and the driver's status information includes: Calculate the safe distance based on the vehicle's real-time speed, the safe speed, and the relative relationship between the vehicle and obstacles in the curve; Based on the vehicle's real-time speed, the safe speed, and the safe distance, a graded warning instruction is generated; Wherein, the safe distance Calculated using the following formula: ; in, For the vehicle's speed, This is the maximum deceleration of the vehicle. For the speed of the obstacle, To decelerate the obstacle, For system delay time, This refers to the driver's reaction time. For environmental dynamics, This represents the driver's dynamic state coefficient.

[0011] Furthermore, the calculation of the safe distance based on the vehicle's real-time speed, the safe speed, and the relative relationship between the vehicle and obstacles within the curve includes: When it is determined that there are no dynamic obstacles in the curve, the target for calculating the safe distance is set as the road boundary of the curve; Based on the vehicle's real-time speed, deceleration, environmental dynamic coefficient, and driver state dynamic coefficient, calculate the anti-exit warning distance between the vehicle and the road boundary; Based on the comparison between the anti-exit warning distance and the actual distance of the vehicle to the road boundary, a corresponding warning command is generated.

[0012] Furthermore, the environmental dynamic coefficient Calculated using the following formula: ; in, For visibility, Let be the duration of rainfall, and be the attenuation constant. Road surface wear; The driver's state dynamic coefficient Calculated using the following formula: ; in, For real-time driving behavior scoring, N is the size of the time window for scoring statistics, and S is the calibration parameter. It is the hyperbolic tangent function.

[0013] Furthermore, obtaining the road parameter information of the curve where the vehicle is located includes: Acquire point cloud data of curved road surfaces collected by roadside sensing devices; Based on the reflection intensity of each point in the road surface point cloud data, the probability density distribution of the reflection intensity is calculated; The current adhesion coefficient of the curved road surface is determined based on the probability density distribution of the reflection intensity. ; Wherein, the current adhesion coefficient Calculated using the following formula: ; in, The reference adhesion coefficient is k, and the adjustment coefficient is k. The baseline probability density; Let I be the probability density weighting function for the reflected intensity I, and its calculation formula is: ; in Let j be the probability density of the reflection intensity at the j-th point in the point cloud. Let be the Gaussian weight of the j-th point, and n be the number of points; Calculated using the following formula: .

[0014] Furthermore, the step of generating a graded warning instruction based on the vehicle's real-time speed and the safe speed includes: Based on historical accident data of the curve, the warning speed threshold is dynamically calculated. ; Compare the vehicle's real-time speed v with the aforementioned safe speed. and the aforementioned warning speed threshold Compare; If v Generate the first-level prompt message; like <v Generate a second-level early warning message; If v> It generates a third-level alarm message and triggers the vehicle's active braking intervention; Among them, the warning speed threshold Calculated using the following formula: in, This represents the historical number of accidents along the curve. The total number of times the vehicle passes through the curve is denoted as , and the sensitivity coefficient is denoted as .

[0015] Secondly, the present invention also provides a cornering driving safety warning system, applied to an in-vehicle terminal, the system comprising: The information acquisition module is used to acquire the real-time driving status information of the vehicle, the road parameter information of the curve where the vehicle is located, and the environmental status information; the environmental status information includes at least one or more of the following: wind speed, visibility, duration of rainfall, and road surface wear. The vehicle speed calculation module is used to calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over, based on the road parameter information of the curve; it is also used to acquire the real-time load information of the vehicle, and determine the wind speed correction coefficient and the load correction coefficient based on the wind speed and the real-time load information, respectively; and calculate the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient and the load correction coefficient. The early warning decision module is used to determine and execute an early warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information.

[0016] Thirdly, the present invention also provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the aforementioned method for warning of safe driving on curves.

[0017] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned method for warning of safe driving on curves.

[0018] Compared with the prior art, the present invention has the following beneficial technical effects: The present invention provides a method, system, device and storage medium for cornering driving safety warning. Based on the vehicle-road cooperative mode, it can deeply integrate driver characteristics, real-time vehicle status and multi-dimensional environmental information, and can dynamically learn and optimize based on historical big data to form a cornering driving safety warning system.

[0019] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute a limitation thereof.

[0021] Figure 1 This is a flowchart of a method for early warning of driving safety on curves according to the present invention; Figure 2 This is a structural diagram of a curve driving safety early warning system according to the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Example 1 This embodiment provides a method for road safety warning when driving on curves, applied to an in-vehicle terminal. The method includes the following steps P1-P5: P1. Obtain the real-time driving status information of this vehicle, the road parameter information of the curve where this vehicle is located, and the environmental status information; the environmental status information includes at least one or more of the following: wind speed, visibility, duration of rainfall, and road surface wear. P2. Based on the road parameter information of the curve, calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over in the curve; P3. Obtain the real-time load information of the vehicle, and determine the wind speed correction coefficient and load correction coefficient respectively based on the wind speed and the real-time load information; P4. The safe speed of the vehicle in the curve is calculated based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient. P5. Determine and execute a warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information.

[0024] The core architecture of this solution is a closed-loop process of perception, computation, decision-making, and execution. It integrates driver status, environmental status (wind speed, visibility, etc.) with traditional vehicle status and road parameters, clearly defining the endpoint for determining and executing early warning strategies, and covering the entire process from early warning to active control. By introducing multi-dimensional real-time data for fusion decision-making, it fundamentally solves the problems of false alarms and missed alarms caused by the rigidity of models in traditional systems, laying the foundation for achieving accurate and acceptable active safety intervention.

[0025] Vehicle status (speed, acceleration) can be acquired through onboard sensors (CAN bus, IMU, GPS); curve parameters (radius, cross slope) and amplified environmental information (provided by roadside temperature, humidity, wind speed, and visibility sensors) can be acquired through roadside units (RSU) and 5G-V2X communication; driver status can be indirectly assessed through onboard DMS cameras or by fusing driving behavior (steering wheel angle fluctuations, braking frequency). Multi-source data is integrated through an onboard computing unit (or in collaboration with edge cloud) to execute the specific algorithms described in subsequent claims. Based on the calculation results, graded warnings (visual, auditory) are issued through the onboard human-machine interface (HMI), or control commands are sent via the vehicle controller (VCU) through the CAN bus to execute active braking or steering assistance.

[0026] In this embodiment, calculating the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient includes: The safe vehicle speed is calculated using the following formula. : ; in, This is the first critical velocity. This is the second critical velocity; The wind speed correction factor is calculated using the following formula: ; in Real-time wind speed can be provided by roadside weather stations or vehicle-mounted sensors; The reference wind speed (a calibration value, such as 20 m / s) quantifies the impact of crosswinds on the lateral stability of vehicles; the higher the wind speed, the lower the lower limit of safe vehicle speed.

[0027] The load correction factor is calculated using the following formula: ; Where L is the real-time load of the vehicle, which can be estimated by a suspension height sensor or a load sensor; The load is the rated load; this formula reflects the negative impact of the change in the center of gravity caused by the increase in load on the rollover threshold.

[0028] Based on the principles of vehicle dynamics, crosswinds increase lateral forces, affecting the sideslip threshold; increased load raises the center of gravity, reducing the rollover threshold. By using two correction coefficients, these two key factors, which are often overlooked in existing curve warning systems, are quantitatively integrated. In strong winds or when trucks are fully loaded, the system can automatically calculate a more conservative and realistic safe speed, improving the accuracy and safety of the warning from the source.

[0029] In this embodiment, determining the warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information includes: P51. Calculate the safe distance based on the vehicle's real-time speed, the safe speed, and the relative relationship between the vehicle and obstacles in the curve; P52. Generate graded warning commands based on the vehicle's real-time speed, the safe speed, and the safe distance; Wherein, the safe distance Calculated using the following formula: ; in, For the vehicle's speed, This is the maximum deceleration of the vehicle. For the speed of the obstacle, The deceleration of obstacles is obtained from sensors and V2X communication. For system delay time, This refers to the driver's reaction time. For environmental dynamics, This represents the driver's dynamic state coefficient.

[0030] When there are no dynamic obstacles, the system does not stop monitoring distance. Instead, it changes the monitoring target from moving vehicles to static road boundaries (such as guardrails, roadbed edges, and oncoming lane lines), which is more in line with the safety requirements of preventing vehicles from going off the road in actual driving.

[0031] When it is determined that there are no dynamic obstacles in the curve, the target for calculating the safe distance is set as the road boundary of the curve; based on the vehicle's real-time speed, deceleration, environmental dynamic coefficient, and driver state dynamic coefficient, the anti-exit warning distance between the vehicle and the road boundary is calculated; based on the comparison between the anti-exit warning distance and the actual distance from the vehicle to the road boundary, a corresponding warning instruction is generated.

[0032] Specifically, the perception and acquisition of road boundaries can be based on high-precision maps and positioning, roadside perception, or vehicle-mounted perception. When the roadside and vehicle-mounted systems confirm that there are no dynamic obstacles within the curve, model degradation is triggered, shifting from preventing collisions with moving obstacles to preventing vehicles from going off the road due to excessive speed, understeering, or other reasons. At this point, the original safety distance formula... In and The value is zero (because the road boundary is stationary), and the formula degenerates into: ; The required safe distance to prevent the vehicle from going out of bounds is, in physical terms, the distance the vehicle travels during the time of driver reaction and system delay, plus the stopping distance after braking begins.

[0033] Driver styles can be categorized as cautious, slightly aggressive, aggressive, slow, and normal. Different drivers have different reaction times, which can be freely set. The initial values ​​for driver styles and corresponding driver reaction times are shown in Table 1, and can be further adjusted... Dynamic adjustment.

[0034] Table 1 Driver Style and Reaction Time Braking efficiency and driver reaction are not static; It dynamically reflects the real-time impact of road surface friction, visibility, and other factors on braking distance; It dynamically reflects the real-time impact of driver focus and fatigue on reaction time; enabling... It can accurately match the real risk at the current moment and realize the soft adjustment of the warning threshold. Under the premise of ensuring safety, it can reduce unnecessary aggressive warnings (such as in good weather or when the driver is focused) and improve the system's acceptance.

[0035] In this embodiment, the environmental dynamic coefficient Calculated using the following formula: ; in, Visibility can be obtained from roadside visibility meters or camera image analysis; The duration of rainfall can be obtained from rainfall sensor timing or weather forecast information; is the attenuation constant. For road surface wear, it can be estimated in real time based on lidar, and the road surface wear condition can be assessed by point cloud reflection intensity and texture features.

[0036] The visibility factor directly affects the risk perception distance; the continuous rainfall factor (exponential decay) simulates the dynamic process of the road surface gradually becoming slippery and the coefficient of adhesion decreasing; and the road wear factor quantifies the condition of road infrastructure. The technical effect is to achieve a refined and proactive assessment of environmental risks; for example, the system gradually increases its alertness when it first starts raining, rather than suddenly changing its strategy when the coefficient of adhesion drops sharply, which aligns with the driver's cognitive patterns.

[0037] In this embodiment, the driver state dynamic coefficient Calculated using the following formula: ; in, Real-time driving behavior scoring can be calculated based on indicators such as steering wheel angle entropy, lateral acceleration variance, and following distance; N is the size of the time window for scoring statistics, and S is a calibration parameter. It is the hyperbolic tangent function.

[0038] A hyperbolic tangent function is used to smoothly normalize the driving behavior score, mapping the overall score to the [0,1] interval, and then adjusting it to the [0.25,1.0] range. The S-shaped characteristic of the hyperbolic tangent function ensures that when the score changes in the middle region, Slow changes (avoiding frequent fluctuations); when scores are extremely high (aggressive) or extremely low (slow), Approaching the boundary value provides sufficient protection margin; achieving smooth and stable adaptation to driver style, it can capture significant changes in driving style while filtering out occasional minor operational fluctuations, making the system intervention appear smarter and more human-like.

[0039] In this embodiment, obtaining the road parameter information of the curve where the vehicle is located includes: Acquire point cloud data of curved road surfaces collected by roadside sensing devices; Based on the reflection intensity of each point in the road surface point cloud data, the probability density distribution of the reflection intensity is calculated; The current adhesion coefficient of the curved road surface is determined based on the probability density distribution of the reflection intensity. ; Wherein, the current adhesion coefficient Calculated using the following formula: ; in, The reference adhesion coefficient is k, and the adjustment coefficient is k. The baseline probability density; Let I be the probability density weighting function for the reflected intensity I, and its calculation formula is: ; in Let j be the probability density of the reflection intensity at the j-th point in the point cloud. Let be the Gaussian weight of the j-th point, and n be the number of points; Calculated using the following formula: .

[0040] By leveraging the strong correlation between point cloud reflection intensity and road surface microtexture (which determines the adhesion coefficient), and by using Gaussian weighting to focus on the local area that the vehicle is about to pass through, and outputting continuous values, the system can distinguish between dry asphalt and polished dry asphalt, states that are difficult to differentiate using lookup tables, thus providing more accurate underlying parameters for vehicle dynamics control.

[0041] In this embodiment, generating a graded warning instruction based on the vehicle's real-time speed and the safe speed includes: P521. Based on the historical accident data of the aforementioned curve, dynamically calculate the warning speed threshold. ; P522. Compare the vehicle's real-time speed v with the stated safe speed. and the aforementioned warning speed threshold Compare; If v Generate the first-level prompt message; like <v Generate a second-level early warning message; If v> It generates a third-level alarm message and triggers the vehicle's active braking intervention; Among them, the warning speed threshold Calculated using the following formula: ; in, This represents the historical number of accidents along the curve. The total number of times a vehicle passes through the curve can be statistically analyzed over a long period by the roadside RSU and stored in the edge cloud or central cloud, and then sent out when a vehicle approaches; the sensitivity coefficient controls the weight of historical data on the threshold.

[0042] Example 2 This embodiment provides a curve driving safety warning system applied to an in-vehicle terminal. The system includes: The information acquisition module 100 is used to acquire the real-time driving status information of the vehicle, the road parameter information of the curve where the vehicle is located, and the environmental status information; the environmental status information includes at least one or more of wind speed, visibility, rainfall duration, and road surface wear. The vehicle speed calculation module 200 is used to calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over, based on the road parameter information of the curve; it is also used to acquire the real-time load information of the vehicle, and determine the wind speed correction coefficient and the load correction coefficient based on the wind speed and the real-time load information, respectively; and calculate the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient and the load correction coefficient. The early warning decision module 300 is used to determine and execute an early warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information.

[0043] The information acquisition module 100 is a distributed data collection and fusion network, implemented collaboratively by roadside and vehicle-mounted components, and can be achieved through the following hardware and software collaborative schemes: I. System Hardware and Data Source Deployment Roadside System: An integrated roadside unit is deployed 150-200 meters in front of the target curve. This unit includes at least: Sensing sensors: LiDAR, millimeter-wave radar, and visual cameras, used to detect obstacles in the curve and acquire road point clouds; Environmental sensors: ultrasonic anemometers, visibility meters, and rain sensors, used to collect environmental status information such as wind speed, visibility, and rainfall duration; Computing and Communication Unit: A built-in edge computing server and roadside unit are used to process raw sensor data and generate structured information.

[0044] Onboard System: Onboard Unit: Used to receive roadside broadcast information; Domain Controller / High-Performance ECU: As the core computing unit, it executes safety algorithms; Vehicle Bus: Used to obtain the vehicle's real-time speed, longitudinal acceleration, steering angle, and real-time load (which can be estimated through the suspension height sensor or the load signal in the CAN bus); Execution and Human-Machine Interface: The vehicle controller is used to receive braking commands; The instrument panel or HUD is used to display warning information; The speaker is used for audible alarms.

[0045] II. Implementation Process of Information Acquisition and Transmission Step A (Roadside Information Acquisition and Broadcast): The roadside edge computing unit fuses multi-sensor data in real time; for example, it uses LiDAR point cloud to extract the road plane through a cloth simulation filtering algorithm, and combines it with pre-stored road design data or real-time calculation to obtain the curve radius and cross slope; at the same time, it detects and tracks all obstacles in the curve to obtain their position, speed, and motion status; finally, it packages all road parameters, environmental status information, and obstacle list, and broadcasts them periodically with specific message sets (such as MAP, SPAT, RSM) through 5G-V2X or C-V2X technology.

[0046] Step B (Vehicle Information Reception and Fusion): The vehicle's OBU receives the aforementioned broadcast information; simultaneously, the vehicle network obtains the vehicle's real-time driving status from the CAN bus; driver status information can be obtained directly (by the DMS driver monitoring camera) or indirectly (by a scoring model based on behavioral characteristics such as steering wheel angle fluctuations and accelerator pedal opening change rates).

[0047] III. Execution of Core Algorithms for Dynamic Computation and Decision Making Calculating safe vehicle speed: The onboard domain controller first calculates the critical speeds to prevent skidding and rollover based on the received information and the vehicle's own parameters (wheelbase, center of gravity height), using vehicle dynamics formulas. Then, it calculates correction coefficients based on the real-time wind speed *w* and vehicle load *L*. Finally, it determines the personalized safe cornering speed under the current conditions according to the formulas.

[0048] Decision-making and execution: The controller compares the vehicle's real-time speed with a safe speed; this comparison is not simply a comparison with a fixed value, but rather a dynamic optimization of the threshold based on historical accident data; based on the comparison results, it generates tiered instructions: If the real-time speed is within a safe range, a friendly warning will be issued via roadside information boards or the vehicle's HMI, indicating that a curve is ahead and warning passengers to be aware.

[0049] If the real-time speed exceeds the safe range, a yellow warning will be triggered in sequence, followed by a red audible and visual alarm, until an active braking request is sent to the VCU in the case of severe speeding, controlling the vehicle to decelerate to below the safe speed.

[0050] IV. Extended Implementation for Scenes with Obstacles When roadside information indicates the presence of a moving obstacle in the curve, the domain controller will invoke the dynamic safe distance model. This model not only includes the speed and deceleration of the vehicle and the target vehicle, but more importantly, it incorporates real-time calculated environmental dynamic coefficients (reflecting road friction and visibility) and driver state dynamic coefficients (reflecting the driver's reaction ability). By calculating the real-time safe distance and comparing it with the actual distance between the two vehicles, the system can make precise decisions from warning to automatic emergency braking.

[0051] Example 3 This embodiment provides an electronic device (such as a vehicle-mounted ECU or a roadside edge server), including a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the curve driving safety warning method described in Embodiment 1.

[0052] Example 4 This embodiment provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the curve driving safety warning method described in Embodiment 1.

[0053] Finally, it should be noted that this article uses specific examples to illustrate the principles and implementation methods of the present invention. The above description of the embodiments is only for the purpose of helping to understand the core ideas of the present invention. Without departing from the principles of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the present invention.

Claims

1. A method for early warning of driving safety on curves, characterized in that, Applied to vehicle-mounted terminals, the method includes: The system acquires real-time driving status information of the vehicle, road parameter information of the curve in which the vehicle is located, and environmental status information; the environmental status information includes at least one or more of the following: wind speed, visibility, duration of rainfall, and road surface wear. Based on the road parameter information of the curve, calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over in the curve. Obtain the real-time load information of the vehicle, and determine the wind speed correction coefficient and load correction coefficient based on the wind speed and the real-time load information, respectively. The safe speed of the vehicle in the curve is calculated based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient. Based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information, a warning strategy is determined and executed.

2. The method according to claim 1, characterized in that, The step of calculating the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient, and the load correction coefficient includes: The safe vehicle speed is calculated using the following formula. : ； in, This is the first critical velocity. This is the second critical velocity; The wind speed correction factor is calculated using the following formula: ； in For real-time wind speed, The reference wind speed; The load correction factor is calculated using the following formula: ； Where L represents the real-time load of the vehicle. This is the rated load.

3. The method according to claim 1, characterized in that, The step of determining a warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information includes: Calculate the safe distance based on the vehicle's real-time speed, the safe speed, and the relative relationship between the vehicle and obstacles in the curve; Based on the vehicle's real-time speed, the safe speed, and the safe distance, a graded warning instruction is generated; Wherein, the safe distance Calculated using the following formula: ； in, For the vehicle's speed, This is the maximum deceleration of the vehicle. For the speed of the obstacle, To decelerate the obstacle, For system delay time, This refers to the driver's reaction time. For environmental dynamics, This represents the driver's dynamic state coefficient.

4. The method according to claim 3, characterized in that, The calculation of the safe distance based on the vehicle's real-time speed, the safe speed, and the relative relationship between the vehicle and obstacles within the curve includes: When it is determined that there are no dynamic obstacles in the curve, the target for calculating the safe distance is set as the road boundary of the curve; Based on the vehicle's real-time speed, deceleration, environmental dynamic coefficient, and driver state dynamic coefficient, calculate the anti-exit warning distance between the vehicle and the road boundary; Based on the comparison between the anti-exit warning distance and the actual distance of the vehicle to the road boundary, a corresponding warning command is generated.

5. The method according to claim 3, characterized in that, The environmental dynamic coefficient Calculated using the following formula: ； in, For visibility, Let be the duration of rainfall, and be the attenuation constant. Road surface wear; The driver's state dynamic coefficient Calculated using the following formula: ； in, For real-time driving behavior scoring, N is the size of the time window for scoring statistics, and S is the calibration parameter. It is the hyperbolic tangent function.

6. The method according to claim 1, characterized in that, The process of obtaining road parameter information for the curve where the vehicle is located includes: Acquire point cloud data of curved road surfaces collected by roadside sensing devices; Based on the reflection intensity of each point in the road surface point cloud data, the probability density distribution of the reflection intensity is calculated; The current adhesion coefficient of the curved road surface is determined based on the probability density distribution of the reflection intensity. ; Wherein, the current adhesion coefficient Calculated using the following formula: ； in, The reference adhesion coefficient is k, and the adjustment coefficient is k. The baseline probability density; Let I be the probability density weighting function for the reflected intensity I, and its calculation formula is: ； in Let j be the probability density of the reflection intensity at the j-th point in the point cloud. Let be the Gaussian weight of the j-th point, and n be the number of points; Calculated using the following formula: 。 7. The method according to claim 1 or 3, characterized in that, The step of generating graded warning commands based on the vehicle's real-time speed and the safe speed includes: Based on historical accident data of the curve, the warning speed threshold is dynamically calculated. ; Compare the vehicle's real-time speed v with the aforementioned safe speed. and the aforementioned warning speed threshold Compare; If v Generate the first-level prompt message; like <v Generate a second-level early warning message; If v> It generates a third-level alarm message and triggers the vehicle's active braking intervention; Among them, the warning speed threshold Calculated using the following formula: ； in, This represents the historical number of accidents along the curve. The total number of times the vehicle passes through the curve is denoted as , and the sensitivity coefficient is denoted as .

8. A curve driving safety warning system, characterized in that, The system, applied to vehicle-mounted terminals, includes: The information acquisition module is used to acquire the real-time driving status information of the vehicle, the road parameter information of the curve where the vehicle is located, and the environmental status information; the environmental status information includes at least one or more of the following: wind speed, visibility, duration of rainfall, and road surface wear. The vehicle speed calculation module is used to calculate the first critical speed at which the vehicle will not skid and the second critical speed at which it will not roll over, based on the road parameter information of the curve; it is also used to acquire the real-time load information of the vehicle, and determine the wind speed correction coefficient and the load correction coefficient based on the wind speed and the real-time load information, respectively; and calculate the safe speed of the vehicle in the curve based on the first critical speed, the second critical speed, the wind speed correction coefficient and the load correction coefficient. The early warning decision module is used to determine and execute an early warning strategy based on the safe vehicle speed, the real-time driving status information of the vehicle, the environmental status information, and the driver status information.

9. An electronic device, characterized in that, include: A processor and a memory, the memory storing a computer program that, when executed by the processor, implements the curve driving safety warning method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the curve driving safety warning method as described in any one of claims 1-7.

Images