Vehicle speed early warning method, device, equipment, medium and product

Abstract

The embodiment of the invention provides a vehicle speed early warning method and device, equipment, a medium and a product, and relates to the field of vehicle auxiliary driving, and the method comprises the steps: calculating the adhesion coefficient of a current road surface in real time through the dynamic parameters of a vehicle and a preset hidden Markov model, and combining the adhesion coefficient with road geometric feature data, thereby achieving the early warning of the vehicle speed. And dynamically calculating the safe vehicle speed adaptive to the current specific vehicle and specific road conditions. Therefore, the early warning threshold value is not fixed any more, but is accurately adjusted in real time according to the friction force of the road surface and the shape of the road. On the basis, layered early warning is triggered according to the actual overspeed severity, so that the conversion from the same warning of all vehicles and road conditions to the accurate and graded warning of specific vehicles under specific road conditions is realized, the suitability of early warning opportunities is improved, and the early warning efficiency is improved. Accurate matching of speed early warning and safety requirements of different driving scenes is achieved, and the actual effect and efficiency of early warning are improved.

Description

Technical Field

[0001] This application relates to the field of vehicle-assisted driving, and more particularly to a method, device, equipment, medium, and product for warning of vehicle speed. Background Technology

[0002] Vehicle speed warning refers to a mechanism that determines whether a vehicle's real-time speed exceeds a safe threshold and triggers a corresponding reminder. During vehicle operation, it is difficult for drivers to determine the safe speed in different scenarios based solely on visual perception and driving experience. Therefore, accurate vehicle speed warning can promptly remind drivers to adjust their speed to adapt to the current driving conditions, which plays an important role in ensuring vehicle driving safety.

[0003] In existing technologies, roadside units are pre-deployed on specific road sections to collect or store static parameters such as the safe speed limit and curvature of the road section in real time, and broadcast them to nearby vehicles via wireless communication. After receiving the information, the vehicles compare it with their own real-time speed, and if speeding is detected, a warning is issued to the driver.

[0004] However, existing technologies suffer from low efficiency in vehicle speed warnings. Current technologies rely on fixed road speed limits or single speed measurement data for warnings, failing to fully integrate multi-dimensional key information affecting vehicle safety. This results in a lack of targeted warning judgments, insufficient adaptability of warning timing, and difficulty in accurately matching the safety needs of different driving scenarios, thus reducing the actual effectiveness and efficiency of speed warnings. Summary of the Invention

[0005] This application provides a method, device, equipment, medium, and product for warning vehicle speed, in order to solve the problem of low efficiency in vehicle speed warning in the prior art.

[0006] In a first aspect, embodiments of this application provide a method for warning vehicle speed, comprising:

[0007] Acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle;

[0008] The road surface adhesion coefficient is calculated based on the multiple vehicle dynamic parameters and a preset hidden Markov model; wherein the road surface adhesion coefficient is used to represent the frictional force between the vehicle's tires and the road surface.

[0009] Based on the multiple road feature data and the road surface adhesion coefficient, a safe vehicle speed is calculated; wherein, the safe vehicle speed refers to the maximum permissible speed at which the vehicle can safely pass through the road, and the safe vehicle speed is used to compare with the real-time speed of the vehicle to determine whether the vehicle's driving state is safe or unsafe, and the multiple vehicle dynamic parameters include the real-time speed;

[0010] A tiered warning strategy is triggered based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the vehicle speed according to the severity of the speeding.

[0011] In one possible design, the hidden Markov model includes a state transition matrix and a confusion matrix. The calculation of the road adhesion coefficient based on the multiple vehicle dynamic parameters and the preset hidden Markov model includes:

[0012] Based on the multiple vehicle dynamic parameters, the state transition matrix and the confusion matrix are constructed; wherein, the state transition matrix is ​​used to represent the probability of mutual transformation between multiple preset road surface categories, the multiple road surface categories including dry cement road surface, wet and slippery cement road surface, icy and snowy road surface and dirt road surface, and the confusion matrix is ​​used to establish the probability mapping relationship between the multiple road surface categories and the multiple vehicle dynamic parameters;

[0013] Based on the state transition matrix, the confusion matrix, and the multiple vehicle dynamic parameters, state identification is performed to obtain the target road surface category corresponding to the road.

[0014] The road surface adhesion coefficient is determined based on the target road surface category corresponding to the road.

[0015] In one possible design, the step of performing state identification based on the state transition matrix, the confusion matrix, and the plurality of vehicle dynamic parameters to obtain the target road surface category corresponding to the road includes:

[0016] Obtain the vehicle model;

[0017] The state transition matrix and the confusion matrix are adjusted according to the vehicle model to obtain the adjusted state transition matrix and the adjusted confusion matrix;

[0018] Based on the adjusted state transition matrix, the adjusted confusion matrix, and the multiple vehicle dynamic parameters, state identification is performed to obtain the target road surface category corresponding to the road.

[0019] In one possible design, prior to acquiring multiple road feature data and multiple vehicle dynamic parameters, the process further includes:

[0020] The system acquires multiple raw road feature data, multiple vehicle dynamic parameters, and vehicle travel distance. The multiple raw road feature data refers to road data without location compensation, which is used to represent the initial geometric structure and initial shape features of the road. The vehicle travel distance refers to the distance traveled by the vehicle within a preset unit time.

[0021] The multiple original road feature data are obtained by performing position compensation based on the vehicle's travel distance.

[0022] In one possible design, calculating the safe vehicle speed based on the plurality of road feature data and the road surface adhesion coefficient includes:

[0023] Based on the multiple road feature data, the radius of curvature of the curve is calculated; wherein, the radius of curvature of the curve is used to represent the degree of curvature of the curve in the road;

[0024] The safe vehicle speed is calculated based on the radius of curvature of the curve, the road surface adhesion coefficient, and the preset gravitational acceleration.

[0025] In one possible design, triggering a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe vehicle speed includes:

[0026] In response to the difference being less than a preset first threshold, a first-level warning strategy is triggered; wherein, the first-level warning strategy refers to generating and playing a voice prompt to remind the driver to slow down;

[0027] In response to the difference being greater than the first threshold and less than a preset second threshold, a second-level warning strategy is triggered; wherein the first threshold is less than the second threshold, and the second-level warning strategy refers to causing the warning icon in the vehicle's dashboard to flash to remind the driver to slow down.

[0028] Secondly, embodiments of this application provide a vehicle speed warning device, comprising:

[0029] The first acquisition module is used to acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle;

[0030] The first calculation module is used to calculate the road surface adhesion coefficient based on the multiple vehicle dynamic parameters and a preset hidden Markov model; wherein the road surface adhesion coefficient is used to represent the friction force between the vehicle's tires and the road surface.

[0031] The second calculation module is used to calculate a safe vehicle speed based on the multiple road feature data and the road surface adhesion coefficient; wherein, the safe vehicle speed refers to the maximum permissible speed at which the vehicle can safely pass through the road, and the safe vehicle speed is used to compare with the real-time speed of the vehicle to determine whether the vehicle's driving state is safe or unsafe, and the multiple vehicle dynamic parameters include the real-time speed;

[0032] The triggering module is used to trigger a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the vehicle speed according to the severity of the vehicle speeding.

[0033] In one possible design, the hidden Markov model includes a state transition matrix and a confusion matrix, and the first computation module includes:

[0034] The construction unit is used to construct the state transition matrix and the confusion matrix based on the multiple vehicle dynamic parameters; wherein, the state transition matrix is ​​used to represent the probability of mutual transformation between multiple preset road surface categories, the multiple road surface categories including dry cement road surface, wet and slippery cement road surface, icy and snowy road surface and dirt road surface, and the confusion matrix is ​​used to establish the probability mapping relationship between the multiple road surface categories and the multiple vehicle dynamic parameters;

[0035] The identification unit is used to perform state identification based on the state transition matrix, the confusion matrix, and the multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road.

[0036] The determining unit is used to determine the road surface adhesion coefficient according to the target road surface category corresponding to the road.

[0037] In one possible design, the identification unit includes:

[0038] A component for obtaining the vehicle model;

[0039] An adjustment component is used to adjust the state transition matrix and the confusion matrix according to the vehicle model to obtain the adjusted state transition matrix and the adjusted confusion matrix;

[0040] The identification component is used to perform state identification based on the adjusted state transition matrix, the adjusted confusion matrix, and the multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road.

[0041] In one possible design, the vehicle speed warning device further includes:

[0042] The second acquisition module is used to acquire multiple original road feature data, multiple vehicle dynamic parameters, and vehicle travel distance; wherein, the multiple original road feature data refers to road data without position compensation, the multiple original road feature data is used to represent the initial geometric structure and initial shape features of the road, and the vehicle travel distance refers to the travel distance of the vehicle in a preset unit time.

[0043] The compensation module is used to perform position compensation on the multiple original road feature data based on the vehicle's travel distance to obtain the multiple road feature data.

[0044] In one possible design, the second computing module includes:

[0045] The first calculation unit is used to calculate the radius of curvature of a curve based on the plurality of road feature data; wherein the radius of curvature of the curve is used to represent the degree of curvature of the curve in the road;

[0046] The second calculation unit is used to calculate the safe vehicle speed based on the curve radius, the road surface adhesion coefficient, and the preset gravitational acceleration.

[0047] In one possible design, the triggering module includes:

[0048] The first triggering unit is configured to trigger a first-level warning strategy in response to the difference being less than a preset first threshold; wherein, the first-level warning strategy refers to generating and playing a voice prompt to remind the driver to slow down;

[0049] The second triggering unit is used to trigger a second-level warning strategy in response to the difference being greater than the first threshold and less than a preset second threshold; wherein the first threshold is less than the second threshold, and the second-level warning strategy refers to causing the warning icon in the vehicle's dashboard to flash to remind the driver to slow down.

[0050] Thirdly, embodiments of this application provide an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0051] The memory stores computer-executed instructions;

[0052] When the processor executes computer execution instructions stored in the memory, it is used to implement the vehicle speed warning method as described in any of the first aspects.

[0053] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the vehicle speed warning method as described in any of the first aspects.

[0054] Fifthly, embodiments of this application provide a computer program product, including a computer program, which, when executed by a processor, is used to implement the vehicle speed warning method as described in any of the first aspects.

[0055] This application provides a vehicle speed warning method, device, equipment, medium, and product. By utilizing the vehicle's dynamic parameters and a preset Hidden Markov Model, it calculates the current road surface adhesion coefficient in real time and combines it with road geometric feature data to dynamically calculate a safe speed suitable for the specific vehicle and road conditions. This makes the warning threshold no longer fixed but adjusted in real time and precisely according to road friction and road shape. Based on this, tiered warnings are triggered according to the severity of the actual speeding, thus realizing a shift from issuing the same warning to all vehicles and road conditions to issuing precise, tiered warnings to specific vehicles under specific road conditions. This improves the adaptability of warning timing, achieves precise matching of speed warnings with the safety requirements of different driving scenarios, and enhances the actual effectiveness and efficiency of the warning system. Attached Figure Description

[0056] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0057] Figure 1 A schematic diagram illustrating an application scenario of the vehicle speed warning method provided in this application embodiment;

[0058] Figure 2 Flowchart of the vehicle speed warning method provided in the embodiments of this application Figure 1 ;

[0059] Figure 3 Flowchart of the vehicle speed warning method provided in the embodiments of this application Figure 2 ;

[0060] Figure 4 A flowchart of the overall curve speed warning method provided in the embodiments of this application;

[0061] Figure 5 This is a structural diagram of a curve speed warning system provided in an embodiment of this application;

[0062] Figure 6 A flowchart for building a road surface feature recognition model provided in this application embodiment;

[0063] Figure 7 A schematic diagram of the structure of the vehicle speed warning device provided in the embodiments of this application;

[0064] Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application.

[0065] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0066] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0067] In the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with substantially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that "first" and "second" do not necessarily imply difference. It should be noted that in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner. In the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more.

[0068] It should be noted that the phrase "at...time" in the embodiments of this application can refer to the instant at which a certain situation occurs, or to a period of time after the occurrence of a certain situation. The embodiments of this application do not specifically limit this. In addition, the vehicle speed warning method, device, equipment, medium, and product provided in the embodiments of this application are only examples. A vehicle speed warning method, device, equipment, medium, and product may also include more or less content.

[0069] To facilitate a clear description of the technical solutions in the embodiments of this application, some terms and technologies involved in the embodiments of this application will be briefly introduced below:

[0070] Hidden Markov Models (HMMs) are time-series dynamic models primarily used for modeling and analyzing sequence data containing hidden states. The core logic of this model is that a system exists with several unobservable hidden states. These states transition according to the rules of a Markov chain, and each hidden state corresponds to a directly observable output, with the observation only related to the current hidden state. It typically consists of five core parts: a set of hidden states, a set of observed states, a state transition probability matrix, an observation probability matrix, and an initial state probability distribution. Through core algorithms such as the forward algorithm, backward algorithm, and Viterbi algorithm, it can achieve functions such as hidden state inference, model parameter training, and observed sequence prediction.

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0072] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0073] To clearly understand the technical solution of this application, the existing technology solutions will first be described in detail. Vehicle speed warning refers to a mechanism that determines whether the real-time speed of a vehicle exceeds a safety threshold and triggers a corresponding reminder. During vehicle operation, accurate vehicle speed warning can promptly remind the driver to adjust the vehicle speed to adapt to the current driving conditions, which plays an important role in ensuring vehicle driving safety.

[0074] In existing technologies, roadside units are pre-deployed on specific road sections to collect or store static parameters such as safe speed limits and curvature in real time, and broadcast them to nearby vehicles. Upon receiving the information, vehicles compare it to their real-time speed; if speeding is detected, a warning is issued to the driver. However, existing technologies rely on fixed road section speed limits or single speed measurement data for warnings, failing to fully integrate multi-dimensional key information affecting vehicle safety. This results in a lack of targeted warning judgments, insufficient adaptability to warning timing, and difficulty in accurately matching the safety needs of different driving scenarios, thus reducing the actual effectiveness and efficiency of speed warnings. Therefore, existing technologies suffer from low efficiency in vehicle speed warnings.

[0075] Therefore, addressing the low efficiency of vehicle speed warning in existing technologies, this research found that to solve this problem, a personalized warning mechanism adapted to specific driving scenarios can be constructed by integrating multi-dimensional real-time information on vehicle dynamic behavior and road environment: ① It can integrate multi-dimensional key information such as road geometry features, real-time traffic conditions, environmental weather conditions, and vehicle operating status, establishing a standardized processing and correlation mechanism for multi-source data. Based on the integrated complete information, it dynamically calculates the safe driving threshold adapted to the current scenario, so that the warning judgment is no longer limited to fixed speed limits, but rather conforms to actual road conditions and vehicle status. ② It can construct an intelligent decision-making model that integrates real-time perception data and historical driving data to achieve early prediction of vehicle driving risks. The model can automatically learn safe driving patterns in different scenarios, analyze the risk probability of the vehicle's current state and future driving path in real time, and upgrade the warning from passive response to proactive prevention. ③ It can divide differentiated warning scenarios based on different road types, weather conditions, traffic flow, and vehicle operating status, formulate appropriate warning triggering rules and response mechanisms for each scenario, and set tiered warning methods according to the severity of vehicle speeding to reduce the interference of ineffective warnings on drivers.

[0076] Specifically, a dynamic early warning system integrating multi-source information can be constructed. This system integrates key information from multiple dimensions, such as road geometry, real-time traffic conditions, environmental weather conditions, and vehicle operating status. Through intelligent algorithm models, various data are comprehensively analyzed to dynamically calculate safe driving thresholds adapted to the current specific driving scenario. Then, based on the degree of deviation between the vehicle's real-time driving status and this threshold, a graded early warning response is triggered. This improves the pertinence and timing adaptability of early warning judgments, achieves precise matching with the safety requirements of different driving scenarios, and enhances the efficiency and effectiveness of vehicle speed warnings.

[0077] This application discloses a vehicle speed warning method, device, equipment, medium, and product. By utilizing the vehicle's own dynamic parameters and a preset hidden Markov model, it calculates the current road surface adhesion coefficient in real time and combines it with road geometric feature data to dynamically calculate a safe speed suitable for the specific vehicle and road surface conditions. This makes the warning threshold no longer fixed but adjusted in real time and accurately according to road surface friction and road shape. Based on this, tiered warnings are triggered according to the severity of the actual speeding, thus realizing a shift from issuing the same warning to all vehicles and road conditions to issuing precise, tiered warnings to specific vehicles under specific road conditions. This improves the adaptability of warning timing, achieves precise matching of speed warnings with the safety requirements of different driving scenarios, and enhances the actual effectiveness and efficiency of the warning system.

[0078] Based on the above-mentioned inventive discovery, the technical solution of this application is proposed.

[0079] The following describes the application scenarios of the vehicle speed warning method provided in the embodiments of the present invention. Figure 1 This is a schematic diagram illustrating an application scenario of the vehicle speed warning method provided in this application embodiment. For example... Figure 1 As shown, this application scenario includes a mobile terminal 101 and a server 102. The mobile terminal 101 collects multiple road feature data and multiple vehicle dynamic parameters, and sends the multiple road feature data and multiple vehicle dynamic parameters to the server 102. The server 102 calculates the road surface adhesion coefficient based on the multiple vehicle dynamic parameters and a preset hidden Markov model. The server 102 calculates the safe speed based on the multiple road feature data and the road surface adhesion coefficient. The server 102 triggers a hierarchical warning strategy based on the numerical range of the difference between the real-time speed and the safe speed.

[0080] The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0081] Figure 2 Flowchart of the vehicle speed warning method provided in the embodiments of this application Figure 1 .like Figure 2 As shown, in this embodiment, the execution entity of this invention is a server. Therefore, the vehicle speed warning method provided in this embodiment includes the following steps:

[0082] S201. Acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle.

[0083] Specifically, vehicle-mounted sensors can collect real-time data on vehicle driving status, while a wireless communication module receives road structure information broadcast from roadside units and retrieves pre-stored road shape data from high-precision maps. By integrating this multi-channel collected or acquired information, multiple road feature data and vehicle dynamic parameters are obtained. This step provides comprehensive and accurate foundational data for subsequent calculations of safe driving parameters adapted to the current driving scenario, ensuring that subsequent operations are based on complete road and vehicle driving status information, thus providing a data foundation for accurate vehicle speed warnings.

[0084] S202. Calculate the road surface adhesion coefficient based on multiple vehicle dynamic parameters and a preset hidden Markov model; whereby the road surface adhesion coefficient is used to represent the frictional force between the vehicle's tires and the road surface.

[0085] Specifically, the collected vehicle dynamic parameters are first preprocessed to extract effective feature parameters. Based on preset road surface categories such as dry cement road, wet cement road, icy and snowy road, and dirt road, a state transition matrix and a confusion matrix are constructed using the effective feature parameters. The state transition matrix corresponds to the probability of transitions between different road surface categories, and the confusion matrix corresponds to the probability mapping relationship between different road surface categories and multiple vehicle dynamic parameters. The state transition matrix, confusion matrix, and multiple vehicle dynamic parameters are then input into a preset Hidden Markov Model for state recognition to determine the target road surface category corresponding to the current road. Finally, based on the preset correspondence between the target road surface category and the road surface adhesion coefficient, the road surface adhesion coefficient of the current road is matched and obtained. This step is used to quantitatively characterize the friction state between the vehicle tires and the current road surface, providing accurate road condition parameters for subsequent calculation of safe speed based on road feature data, ensuring that the calculated safe speed closely matches actual road conditions.

[0086] The road surface adhesion coefficient is a parameter that reflects the gripping ability between the road surface and the vehicle tires. It directly reflects the anti-skid performance of the road surface. Its value is strongly correlated with the road surface type, specifically corresponding to the anti-skid characteristics of different scenarios such as dry cement road surface, wet and slippery cement road surface, icy and snowy road surface, and dirt road surface. This coefficient needs to be calculated in real time through an estimation model based on Hidden Markov Model. It directly affects the balance between the vehicle's centrifugal force and the road surface grip force, thereby determining the stability of cornering and preventing the vehicle from losing stability and overturning due to insufficient grip.

[0087] S203. Calculate the safe speed based on multiple road feature data and road surface adhesion coefficient; whereby the safe speed refers to the maximum permissible speed at which a vehicle can safely pass through the road. The safe speed is used to compare with the vehicle's real-time speed to determine whether the vehicle's driving status is safe or unsafe. Multiple vehicle dynamic parameters include real-time speed.

[0088] Specifically, a multi-parameter fusion model for calculating safe vehicle speed can be constructed. This model uses multiple road feature data, such as road curvature, slope, curve radius, and lane width, as well as the road surface adhesion coefficient, as core input parameters. Simultaneously, it incorporates the vehicle's own braking system performance parameters and substitutes them into a pre-defined vehicle dynamics formula to calculate the vehicle's safe speed under current road conditions. This step determines the maximum permissible speed for a vehicle to safely pass through the current road, providing a clear safety reference standard for subsequent comparisons of real-time vehicle speeds.

[0089] S204. Trigger a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the speed according to the severity of the vehicle speeding.

[0090] Specifically, the difference between the vehicle's real-time speed and the safe speed can be calculated first. A first threshold and a second threshold are preset, with the first threshold being less than the second. Then, the range of this difference is determined. If the difference is less than the preset first threshold, a voice prompt is generated and played. If the difference is greater than the first threshold but less than the preset second threshold, a warning icon on the vehicle's dashboard flashes, triggering the corresponding tiered warning strategy. This step uses differentiated alerts based on the degree to which the vehicle's real-time speed exceeds the safe speed, delivering a deceleration signal to the driver and providing clear guidance for speed adjustment.

[0091] Vehicle speed warning methods can be widely applied to vehicle active safety systems, advanced driver assistance systems, and vehicle-road cooperative scenarios. Especially in complex driving environments with high speed sensitivity and variable road conditions, such as mountain curves, wet and slippery roads, urban expressway ramps, and highway construction zones, this method can provide accurate speed guidance and graded warnings based on real-time calculated personalized safety thresholds. This effectively assists drivers or autonomous driving systems in avoiding risks such as skidding and loss of control caused by improper speed, thereby improving driving safety under various road conditions.

[0092] This embodiment provides a vehicle speed warning method that utilizes the vehicle's dynamic parameters and a pre-set Hidden Markov Model to calculate the current road surface adhesion coefficient in real time. This coefficient is then combined with road geometric feature data to calculate a safe speed suitable for the specific vehicle and road conditions. This allows the warning threshold to be adjusted precisely based on road friction and road shape, rather than being fixed. Furthermore, tiered warnings are triggered according to the severity of the actual speeding, thus shifting from issuing the same warning to all vehicles and road conditions to providing precise, tiered warnings to specific vehicles under specific road conditions. This improves the adaptability of warning timing, achieves accurate matching of speed warnings with the safety requirements of different driving scenarios, and enhances the actual effectiveness and efficiency of the warning system.

[0093] In one possible design, the Hidden Markov Model includes a state transition matrix and a confusion matrix. S202, based on multiple vehicle dynamic parameters and a pre-defined Hidden Markov Model, calculates the road adhesion coefficient, including:

[0094] S2021. Based on multiple vehicle dynamic parameters, construct a state transition matrix and a confusion matrix. The state transition matrix is ​​used to represent the probability of mutual transformation between multiple preset road surface categories, including dry cement road surface, wet and slippery cement road surface, icy and snowy road surface and dirt road surface. The confusion matrix is ​​used to establish the probability mapping relationship between multiple road surface categories and multiple vehicle dynamic parameters.

[0095] Specifically, the collected vehicle dynamic parameters are first cleaned and feature extracted to filter out effective parameters related to road conditions. Based on four preset road surface categories—dry cement road, wet cement road, icy and snowy road, and dirt road—and combined with historical road surface transition data of vehicle travel, the probability values ​​of transitions between different road surface categories are calculated to construct a state transition matrix. Simultaneously, by analyzing the distribution patterns of vehicle dynamic parameters under each road surface category, a probability mapping relationship between each category and vehicle dynamic parameters is established, thereby constructing a confusion matrix. This step provides the core probability calculation basis for the Hidden Markov Model, clarifying the transition patterns between road surface categories and the correlation between road surface categories and vehicle dynamic parameters, providing standardized matrix parameter support for subsequent road surface condition identification.

[0096] For example, taking a family sedan as a test vehicle, 13 dynamic vehicle parameters were collected in real time via the vehicle's CAN bus and high-precision sensors, including vehicle speed, left front wheel speed, right front wheel speed, left rear wheel speed, right rear wheel speed, lateral acceleration, longitudinal acceleration, yaw rate, engine output torque, steering wheel angle, accelerator pedal position, gear position, and brake pedal position. Then, under four scenarios—dry cement road surface, artificially watered wet cement road surface, low-temperature icy and snowy road surface, and unpaved dirt road surface in the suburbs—the vehicle was driven at different speeds from 20-80 km / h, performing straight-line driving, turning, acceleration, and braking under various conditions. A total of 400,000 sets of valid data were continuously collected over 80 hours. After noise reduction and normalization preprocessing, the switching between the four road surface states was first obtained from all the data. For example... The probability of the transition path is obtained by dividing the number of times the dry cement road surface switches to the wet and slippery cement road surface by the total number of times the dry cement road surface state occurs. The transition probabilities between all states are calculated in this way to initially construct a 4×4 dimensional state transition matrix. Then, the frequency of occurrence of each combination of vehicle dynamic parameters under each road surface state is calculated. For example, the frequency of occurrence of the parameter combination of vehicle speed 50km / h, lateral acceleration 1.5m / s², and right rear wheel speed 49.8km / h under icy and snowy road surface is divided by the total number of observations on icy and snowy road surface to obtain the corresponding output probability. This leads to the initial construction of a 4×M dimensional confusion matrix. Finally, the hidden Markov model can be iteratively trained using the Baum-Welch algorithm to continuously adjust the probability parameters of the two matrices until the generation probability of the observation sequence converges, thus obtaining the optimal state transition matrix and confusion matrix.

[0097] The state transition matrix is ​​a square matrix in the Hidden Markov Model (HMM) that describes the dynamic switching patterns between various road surface categories. Its matrix elements represent the probability that a vehicle will transition to another road surface state at the next moment when it is currently in a certain road surface state. Its core purpose is to quantify the evolution trend of road surface states. The confusion matrix is ​​a matrix in the HMM that establishes the correspondence between the hidden road surface state and the observed values ​​of vehicle dynamic parameters. Its matrix elements represent the probability that a specific combination of vehicle dynamic parameters is observed when the vehicle is in a certain road surface state. Its core purpose is to infer the true road surface state from the collected parameters.

[0098] S2022. Based on the state transition matrix, confusion matrix and multiple vehicle dynamic parameters, state recognition is performed to obtain the target road surface category corresponding to the road.

[0099] Specifically, multiple preprocessed vehicle dynamic parameters can be used as an observation sequence and substituted into a Hidden Markov Model (HMM) containing a state transition matrix and a confusion matrix. Using a pre-defined state inference algorithm, combined with the transition probabilities between road surface categories in the state transition matrix and the probability mapping relationship between road surface categories and vehicle dynamic parameters in the confusion matrix, the current road surface state is inferred sequentially. The road surface category with the highest probability is output as the target road surface category. This step clarifies the specific road surface type, providing a category basis for subsequent matching of the corresponding road surface adhesion coefficient based on the road surface category, ensuring that the determined road surface adhesion coefficient accurately reflects the actual road surface conditions.

[0100] For example, a family sedan with a completed model training can be selected as an example. During the vehicle's operation, multiple vehicle dynamic parameters are collected in real time via the CAN bus and high-precision sensors. The collection frequency is set to 100Hz. The collected raw parameters are denoised and normalized through a sliding window to transform them into an observation sequence that meets the input requirements of the Hidden Markov Model. Then, this observation sequence, along with the trained state transition matrix (4×4-dimensional, quantifying the transition probabilities between four road surface states) and confusion matrix (4×M-dimensional, quantifying the corresponding probabilities of each road surface state and parameter combination), are input into the model. The Viterbi algorithm is used to calculate the optimal hidden state sequence corresponding to the current observation sequence. By maximizing the product of the current road surface state probability × state transition probability × observation probability, the category with the highest probability is selected from four candidate categories: dry cement road surface, wet and slippery cement road surface, icy and snowy road surface, and dirt road surface. This category is the target road surface category corresponding to the current road. The latency of the entire recognition process can be controlled within 50ms, meeting the real-time requirements.

[0101] S2023. Determine the road surface adhesion coefficient according to the target road surface category corresponding to the road.

[0102] Specifically, a mapping table can be pre-established, mapping target road surface categories (dry cement road surface, wet cement road surface, icy and snowy road surface, and dirt road surface) to their corresponding road surface adhesion coefficients. This table stores the standard adhesion coefficient values ​​for each category. Once the target road surface category is determined, the mapping table is directly queried to obtain the corresponding road surface adhesion coefficient value. This step quantifies the friction state between the vehicle tires and the current road surface, providing accurate road condition parameters for subsequent calculations of safe speeds based on road feature data, ensuring that the calculated safe speed closely reflects actual road conditions.

[0103] For example, a mapping table between target road surface categories and road surface adhesion coefficients can be pre-established. Dry cement road surfaces correspond to adhesion coefficients of 0.75-0.85, wet and slippery cement road surfaces correspond to 0.45-0.55, icy and snowy road surfaces correspond to 0.15-0.25, and dirt road surfaces correspond to 0.35-0.45. The mapping table is stored locally in the vehicle control system. When the target road surface category is identified, the system quickly matches the corresponding road surface adhesion coefficient range by looking up the table. It then makes dynamic fine adjustments based on the current vehicle driving conditions (such as vehicle speed and braking status). For example, when the vehicle speed is higher than 60 km / h on a wet and slippery cement road surface, the coefficient is lowered to 0.42 to ensure that the coefficient matches the actual road surface grip.

[0104] The technical effect of this solution in this embodiment is as follows: By applying a structured Hidden Markov Model (HMM), vehicle dynamic parameters are mapped to specific road surface categories, enabling accurate determination of the road surface adhesion coefficient. A state transition matrix is ​​used to model the temporal pattern of road surface state changes, and a confusion matrix is ​​used to model the probability distribution of vehicle dynamic characteristics under specific road surfaces. The synergy of these two methods allows the system to not only identify the current road surface type but also understand the possibility of state transitions, improving the accuracy of road surface state identification. This lays the foundation for subsequent calculations of safe speeds highly adapted to real-world road conditions and solves the problem of inaccurate warning benchmarks caused by the inability of traditional methods to perceive dynamic changes in the road surface.

[0105] In one possible design, S2022, state recognition is performed based on the state transition matrix, confusion matrix, and multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road, including:

[0106] S20221. Obtain the vehicle model.

[0107] Specifically, the vehicle model can be determined by reading pre-stored vehicle configuration information in the vehicle control system, obtaining the vehicle identification code through the vehicle's on-board diagnostic interface, and then parsing the vehicle model based on the preset correspondence between the vehicle identification code and the vehicle model. Alternatively, the vehicle model information can be received from the user through the on-board interface. This step provides vehicle model-related parameters for subsequent adjustments to the state transition matrix and confusion matrix, ensuring that the matrix parameters are adapted to the characteristics of the current vehicle, and that subsequent road condition recognition operations conform to the actual driving characteristics of the vehicle.

[0108] S20222. Adjust the state transition matrix and confusion matrix according to the vehicle model to obtain the adjusted state transition matrix and the adjusted confusion matrix.

[0109] Specifically, a matrix adjustment coefficient library covering different vehicle models can be pre-established. This library stores adjustment coefficients related to characteristics such as tire grip performance and vehicle weight for each model. Once the vehicle model is identified, the target adjustment coefficient corresponding to the current model is matched from this coefficient library. Then, based on the transition probabilities between various road surface categories in the state transition matrix and the probability mapping relationship between road surface categories and vehicle dynamic parameters in the confusion matrix, the values ​​in the matrices are corrected according to the target adjustment coefficients, resulting in the adjusted state transition matrix and the adjusted confusion matrix. This step is used to adapt the parameters of the state transition matrix and the confusion matrix to the characteristics of the current vehicle model, ensuring that the matrix parameters conform to the actual driving response patterns of the vehicle, providing a calculation basis for subsequent road condition recognition operations.

[0110] For example, when the current vehicle is identified as a heavy-duty truck, a pre-set vehicle model calibration template in the database can be invoked. This template is built based on a large amount of real-vehicle test data, showing that due to its greater weight, stiffer suspension, and deeper tire treads, the dynamic response of this type of vehicle on the same road surface differs from that of a regular passenger car. The system calibrates the model matrix according to the specific parameters in the template: for example, in the state transition matrix, considering that heavy-duty trucks are more sensitive to slippery road surfaces, the probability coefficient for transitioning from slippery cement road surface to icy road surface is increased from 0.15 for passenger cars to 0.23; in the confusion matrix, considering the characteristic that heavy-duty trucks have smaller fluctuations in yaw rate on bumpy dirt roads, the probability of the yaw rate observation value in the low range under the dirt road category is increased from 0.3 to 0.45. Through this element-by-element coefficient correction based on the vehicle's dynamic characteristics, a state transition matrix and confusion matrix optimized specifically for this heavy-duty truck are generated.

[0111] S20223. Based on the adjusted state transition matrix, the adjusted confusion matrix, and multiple vehicle dynamic parameters, state recognition is performed to obtain the target road surface category corresponding to the road.

[0112] Specifically, the collected vehicle dynamic parameters are first filtered, denoised, and feature extracted to obtain standardized observation sequence parameters. The adjusted state transition matrix, the adjusted confusion matrix, and the standardized observation sequence parameters are then input into a pre-defined Hidden Markov Model (HMM). The model's built-in state inference algorithm is invoked, combining the transition probabilities between road surface categories in the adjusted matrix and the probability mapping relationship between road surface categories and vehicle dynamic parameters. Sequence calculations and matching are performed on the current road surface state, and the road surface category with the highest probability value is output as the target road surface category. This step is used to identify the road surface type suitable for the current vehicle model characteristics, providing an accurate category reference for subsequently determining the road surface adhesion coefficient based on the target road surface category, ensuring that the determined road surface adhesion coefficient closely matches the actual matching state between the vehicle and the road surface.

[0113] The technical effect of this solution in this embodiment is as follows: by introducing vehicle model information to adjust the state transition matrix and confusion matrix of the Hidden Markov Model, the personalized accuracy of road adhesion coefficient calculation is improved. Different vehicle models, due to differences in weight, center of gravity, tire characteristics, and suspension systems, will exhibit different dynamic parameter characteristics on the same road surface, making it easy for general models to misjudge. By performing vehicle model adaptation correction on the model, the state recognition process is made more consistent with the actual dynamic response of the current vehicle, ensuring a more accurate mapping relationship from vehicle dynamic parameters to road category inference. This provides personalized input for the basic safe speed calculation, enhancing the universality and accuracy of the entire warning system across different vehicle models.

[0114] In one possible design, S203 calculates the safe vehicle speed based on multiple road characteristic data and the road surface adhesion coefficient, including:

[0115] S2031. Calculate the radius of curvature of the curve based on multiple road feature data; whereby the radius of curvature of the curve is used to represent the degree of curvature of the curve in the road.

[0116] Specifically, the process can begin by extracting a continuous sequence of geometric coordinate points for curved road segments from multiple road feature datasets. A curve fitting algorithm is then used to fit this sequence to an arc, calculating the radius of the fitted arc. Alternatively, the change in tangent angle and the corresponding road segment length can be extracted from the road feature data. The curvature of the curve can be derived using geometric formulas, and the radius of curvature can be calculated based on the conversion relationship between curvature and radius. This step quantifies the degree of curvature in road curves, providing crucial geometric parameters for calculating safe speeds by combining the road surface adhesion coefficient and a preset gravitational acceleration. This ensures that the calculated safe speed is accurate and reflects the actual curvature of the curve.

[0117] S2032. Calculate the safe vehicle speed based on the radius of curvature of the curve, the road surface adhesion coefficient, and the preset gravitational acceleration.

[0118] Specifically, the curve radius of curvature, road surface adhesion coefficient, and preset gravitational acceleration can be used as core parameters. These parameters are then substituted into a preset vehicle curve dynamics calculation formula. Through mathematical operations such as multiplication and square root, the safe speed value of the vehicle under the current curve road surface conditions is obtained. This step is used to determine the maximum permissible speed for the vehicle to safely pass through the current curve segment, providing a clear safety reference standard for subsequent comparisons of real-time vehicle speeds, and enabling the early warning judgment to have a quantitative basis that fits the actual situation of the curve.

[0119] The technical effect of this solution in this embodiment is as follows: the curvature radius of the curve is calculated from multiple road feature data to accurately quantify the curvature of the curve, ensuring that the calculated safe speed conforms to the actual structural characteristics of the current road. Then, the curvature radius of the curve is combined with the road surface adhesion coefficient and gravitational acceleration, which reflect the friction state between the tire and the road surface, to calculate a safe speed that is suitable for the current curve and road surface conditions. This transforms the safe speed from a fixed road section speed limit to a precise threshold that dynamically matches the road structure and road surface conditions, avoiding the problem of fixed speed limits not matching actual road conditions. This provides a reliable speed benchmark for the accurate triggering of subsequent layered early warning strategies.

[0120] In one possible design, S204 triggers a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed, including:

[0121] S2041. In response to the difference being less than a preset first threshold, a first-level warning strategy is triggered; wherein, the first-level warning strategy refers to generating and playing a voice prompt to remind the driver to slow down.

[0122] Specifically, the difference between the vehicle's real-time speed and the safe speed can be calculated first. This difference is then compared to a preset first threshold. If the difference is less than the first threshold, the in-vehicle voice generation module is invoked to generate a voice signal containing a deceleration prompt. This voice signal is then transmitted to the in-vehicle audio system, which plays the prompt. This step is used to deliver a deceleration signal to the driver when the vehicle's real-time speed is only slightly exceeding the safe speed, providing clear speed adjustment guidance through voice prompts, thus meeting the reminder needs for minor speeding scenarios.

[0123] S2042. In response to the difference being greater than a first threshold and less than a preset second threshold, a second-level warning strategy is triggered; wherein, the first threshold is less than the second threshold, and the second-level warning strategy refers to causing the warning icon in the vehicle's dashboard to flash to remind the driver to slow down.

[0124] Specifically, the difference between the vehicle's real-time speed and the safe speed can be calculated first. This difference is then compared to a preset first threshold and a second threshold. When the difference is determined to be greater than the first threshold and less than the second threshold, a warning trigger signal is sent to the vehicle's instrument panel control system. Upon receiving the signal, the control system activates a preset warning icon on the instrument panel, causing it to flash continuously. This step is used to convey a deceleration signal to the driver when the vehicle's real-time speed exceeds the safe speed by a moderate degree, using a flashing icon visual reminder to enhance the warning effect.

[0125] The technical effect of this solution in this embodiment is that by transforming a single warning action into a graded response mechanism bound to the speeding risk level, refined management of warning intervention is achieved. By setting different threshold ranges and matching warning methods of different intensities, the system can provide appropriate warning intensity according to the severity of speeding. This design avoids the problems that may arise from traditional one-size-fits-all warning methods: excessive interference to the driver in cases of slight speeding, or insufficient warning intensity in cases of severe speeding. While ensuring the effectiveness of safety warnings, it also optimizes the human-computer interaction experience.

[0126] Figure 3 Flowchart of the vehicle speed warning method provided in the embodiments of this application Figure 2 In this embodiment, in Figure 2 Based on the provided embodiments, the vehicle speed warning method is further explained. The vehicle speed warning method includes:

[0127] S301. Acquire multiple original road feature data, multiple vehicle dynamic parameters, and vehicle travel distance; wherein, multiple original road feature data refers to road data without location compensation, and multiple original road feature data are used to represent the initial geometric structure and initial shape features of the road, and vehicle travel distance refers to the distance traveled by the vehicle within a preset unit time.

[0128] Specifically, the system can receive uncompensated initial road geometry and shape feature data broadcast by roadside units via a wireless communication module, retrieve pre-stored original road information from a high-precision map, and simultaneously collect vehicle driving status parameters in real time using onboard sensors. The vehicle speed sensor records the driving speed within a preset unit of time, and the driving distance is calculated by combining this with time parameters. Alternatively, the onboard positioning system can obtain displacement data within a unit of time as the driving distance. Integrating this multi-channel information yields multiple sets of original road feature data, multiple vehicle dynamic parameters, and vehicle driving distances. This step provides complete initial data support for subsequent location compensation of the original road feature data, ensuring that the location compensation operation is based on accurate vehicle displacement information and original road data, laying the foundation for obtaining accurate road feature data later.

[0129] In this context, "multiple raw road feature data" refers to unprocessed road geometry information acquired from maps or sensors, which contains static or delayed errors relative to the vehicle's actual position. "Multiple road feature data" refers to road geometry information that, after position compensation correction, precisely corresponds to the vehicle's current real-time position. The difference lies in whether dynamic spatial alignment correction has been performed based on the vehicle's real-time displacement. Raw data serves as static or lagging reference values, while processed data provides accurate input that is matched in real-time to the vehicle's trajectory.

[0130] S302. Based on the vehicle's travel distance, perform position compensation on multiple original road feature data to obtain multiple road feature data.

[0131] Specifically, the initial position coordinates and driving direction of the vehicle can be obtained first through the vehicle positioning system. Combined with the vehicle's travel distance, the displacement of the vehicle within a preset unit of time can be calculated. Then, the initial coordinates of multiple original road feature data points are correlated with the vehicle's displacement to correct the positional information of the original data. Simultaneously, the corrected data is calibrated with reference to the coordinate system of a high-precision map, ensuring that the positional information of the road feature data accurately corresponds to the road segment the vehicle is currently traveling on. This results in multiple position-compensated road feature data points. This step corrects the positional deviation of the original road feature data, ensuring that the road feature data accurately reflects the actual structure and shape of the road segment the vehicle is currently traveling on, providing accurate road information support for subsequent calculations of safe speed based on the road surface adhesion coefficient.

[0132] Position compensation is used to correct the misalignment between road geometry data and the vehicle's real-time position caused by continuous vehicle movement. When the system acquires raw feature data such as curvature and slope of the road ahead from a high-precision map or roadside unit, this data typically corresponds to fixed points in the map coordinate system. However, there is an unavoidable slight delay from data acquisition and transmission to processing by the in-vehicle system. During this extremely short lag time, the vehicle has already traveled a certain distance. Position compensation calculates the precise distance the vehicle has traveled during this processing delay in real time and shifts all coordinate points in the original road feature data by that distance in the opposite direction of the vehicle's travel. This allows key locations in the road feature data, such as the start of curves and points of curvature change, to be accurately matched with the vehicle's current absolute position. This ensures that the road geometry information used for subsequent safe speed calculations completely corresponds to the actual road segment the vehicle is about to enter, eliminating the warning position deviation caused by data lag.

[0133] S303. Acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle.

[0134] S304. Calculate the road surface adhesion coefficient based on multiple vehicle dynamic parameters and a preset hidden Markov model; whereby the road surface adhesion coefficient is used to represent the frictional force between the vehicle's tires and the road surface.

[0135] S305. Calculate the safe speed based on multiple road characteristic data and road surface adhesion coefficient; whereby the safe speed refers to the maximum permissible speed at which a vehicle can safely pass through the road. The safe speed is used to compare with the vehicle's real-time speed to determine whether the vehicle's driving status is safe or unsafe. Multiple vehicle dynamic parameters include real-time speed.

[0136] S306. Trigger a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the speed according to the severity of the vehicle speeding.

[0137] S303-S306 are similar to S201-S204, and will not be described again in this embodiment.

[0138] The technical effect of this solution in this embodiment is as follows: By introducing a dynamic position compensation mechanism based on real-time vehicle speed, the problem of data lag and spatiotemporal misalignment caused by fixed-frequency updates of high-precision maps or sensor data in high-speed driving scenarios is solved. By calculating and compensating for the actual distance traveled by the vehicle during the data acquisition and processing delay, the system can accurately align the original road geometric feature data with the vehicle's instantaneous real position. This step ensures that the road feature data subsequently used to calculate the safe speed is accurate spatial information that is synchronized with the vehicle's current position in real time, thereby avoiding errors in curve recognition, inaccurate calculation of safe distances, and misjudgment of warning timing caused by positional deviations, and improving the reliability of the warning system.

[0139] Figure 4 A flowchart of the overall curve speed warning method provided in the embodiments of this application is shown below. Figure 4 As shown, a curve speed warning method based on road geometry features includes the following steps:

[0140] Step 1: Use a high-precision map to obtain the vehicle's location and the road curvature corresponding to different positions in front of the vehicle.

[0141] Step 2: Calculate the safe speed for the curve ahead.

[0142] Step 3: Determine if the vehicle speed is greater than the safe speed.

[0143] Step 4: If the vehicle is speeding, calculate the safe distance.

[0144] Step 5: If the vehicle is less than the safe distance, the curve speed warning function will proactively issue a warning to remind the driver to slow down. The warning will stop once the vehicle speed falls below the safe speed.

[0145] Figure 5 The structural diagram of the curve speed warning system provided in the embodiments of this application is as follows: Figure 5 As shown, the vehicle's position and the road curvature corresponding to different positions in front of the vehicle are obtained using a high-precision map.

[0146] The high-precision map updates the vehicle's position periodically, and the update distance increases with vehicle speed. The cornering speed warning function continuously calculates the vehicle's travel distance based on speed to compensate for the position sent by the map.

[0147]

[0148] in, This is the location of the vehicle. The location of this vehicle is sent to the high-precision map. The time when the vehicle's location is sent to the high-precision map. v represents the current moment, and v represents the vehicle's real-time speed.

[0149] The high-precision map sends information about the curvature of the road ahead, and the curve speed warning function calculates the distance between the curve curvature position and the vehicle.

[0150]

[0151] in, The distance from the curvature position to the vehicle. This indicates the location of the curve curvature.

[0152] In the second step, calculate the safe speed for the curve ahead.

[0153]

[0154] in, For safe driving speed, The road surface adhesion coefficient, It is the acceleration due to gravity. This refers to the curvature of the curve.

[0155] Figure 6 The flowchart for building a road surface feature recognition model provided in the embodiments of this application is as follows: Figure 6 As shown, a road adhesion coefficient estimation model based on Hidden Markov Model is established to calculate the road adhesion coefficient in real time.

[0156] A Hidden Markov Model (HMM) consists of two sets of random states and three sets of probability matrices, and can be represented as:

[0157]

[0158] in, Let M be the number of random hidden states, A be the number of observables, B be the state transition matrix, and C be the confusion matrix. Let be the initial state probability matrix. It is a Hidden Markov Model.

[0159] The number of hidden states in the model is determined using the Bayesian information criterion. Based on the subjective estimate of the number of hidden states, the optimal solution is determined by adjusting the posterior probability using Bayesian methods.

[0160]

[0161] In the formula, For the model, Let N be the model space, and N be the number of random hidden states. For observation data, This is the set of candidate models corresponding to the number of hidden states N. Given model parameters When, the conditional probability density function of the observed data g.

[0162]

[0163] in, yes The maximum likelihood estimate, For the sample size, This is the computational expression for the Bayesian Information Criterion (BIC).

[0164] Minimize the BIC value This is the optimal model.

[0165]

[0166] in, For the hidden state sequence, for The hidden state sequence at time step For probability, for The observation sequence at any given time.

[0167] Based on the number of hidden states, different features of the observation sequence are extracted, and the following methods are used. The mean clustering algorithm is used for clustering, with the goal of minimizing the squared error. .

[0168]

[0169] in, Let be the center vector of the i-th cluster. Let k be the i-th cluster and k be the number of cluster categories.

[0170] For known observation sequences Perform parameter estimation to maximize the probability of generating the observation sequence.

[0171]

[0172] in, The optimal hidden Markov model is... In a given model Under these conditions, observation sequence The probability of occurrence.

[0173] The optimal observation probability and model parameters can be obtained by continuing until the observation probability converges.

[0174] Using different scene data composed of vehicle motion feature parameters, four different road surface category recognition models were trained: dry cement road surface recognition model, wet and slippery cement road surface recognition model, icy and snowy road surface recognition model, and dirt road surface recognition model.

[0175] The motion characteristic parameters include vehicle speed, left front wheel speed, right front wheel speed, left rear wheel speed, right rear wheel speed, lateral acceleration, longitudinal acceleration, yaw rate, engine output torque, steering wheel angle, accelerator pedal position, gear, and brake pedal position.

[0176] like Figure 5 As shown, the vehicle speed, engine output torque, and accelerator pedal position are obtained through the engine controller; the left front wheel speed, right front wheel speed, left rear wheel speed, and right rear wheel speed, lateral acceleration, longitudinal acceleration, yaw rate, and brake pedal position are obtained through the braking system controller; the steering wheel angle is obtained through the steering gear controller; and the current gear is obtained through the transmission controller.

[0177] Through continuous learning and iteration, the following model parameters are obtained for identifying dry cement pavement, wet and slippery cement pavement, icy and snowy pavement, and dirt pavement: the final random state transition probability matrix A, the output probability transition matrix B, and the probability matrix of the initial random state distribution. .

[0178] By inputting the motion characteristic parameters, the road surface category can be output, and the road adhesion coefficient of the corresponding road surface can be obtained. .

[0179] In the third step, it is determined whether the vehicle speed is greater than the safe speed.

[0180] In the fourth step, if the vehicle is speeding, a safe distance is calculated. .

[0181]

[0182] in, This is the vehicle's maximum deceleration. For safe vehicle speed, v represents the vehicle's current real-time speed.

[0183] Step 5: If the vehicle is less than the safe distance, the curve speed warning function will proactively issue a warning to remind the driver to slow down. The warning will stop once the vehicle speed falls below the safe speed.

[0184] Figure 7 This is a schematic diagram of the vehicle speed warning device provided in an embodiment of this application. Figure 7 As shown, the vehicle speed warning device includes:

[0185] The first acquisition module 701 is used to acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle.

[0186] The first calculation module 702 is used to calculate the road surface adhesion coefficient based on multiple vehicle dynamic parameters and a preset hidden Markov model; wherein the road surface adhesion coefficient is used to represent the friction force between the vehicle's tires and the road surface.

[0187] The second calculation module 703 is used to calculate the safe speed based on multiple road feature data and road surface adhesion coefficient. The safe speed refers to the maximum permissible speed at which a vehicle can safely pass through the road. The safe speed is compared with the real-time speed of the vehicle to determine whether the vehicle's driving status is safe or unsafe. The multiple vehicle dynamic parameters include the real-time speed.

[0188] The trigger module 704 is used to trigger a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the speed according to the severity of the vehicle speeding.

[0189] In one possible design, the Hidden Markov Model includes a state transition matrix and a confusion matrix. The first computation module 702 includes:

[0190] The construction unit is used to construct a state transition matrix and a confusion matrix based on multiple vehicle dynamic parameters. The state transition matrix is ​​used to represent the probability of mutual transformation between multiple preset road surface categories, including dry cement road surface, wet and slippery cement road surface, icy and snowy road surface and dirt road surface. The confusion matrix is ​​used to establish the probability mapping relationship between multiple road surface categories and multiple vehicle dynamic parameters.

[0191] The identification unit is used to identify the state based on the state transition matrix, confusion matrix and multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road.

[0192] The determination unit is used to determine the road surface adhesion coefficient based on the target road surface category corresponding to the road.

[0193] In one possible design, the identification unit includes:

[0194] Get the component used to obtain the vehicle model.

[0195] An adjustment component is used to adjust the state transition matrix and confusion matrix according to the vehicle model, resulting in an adjusted state transition matrix and an adjusted confusion matrix.

[0196] The identification component is used to identify the state based on the adjusted state transition matrix, the adjusted confusion matrix, and multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road.

[0197] In one possible design, the vehicle speed warning device also includes:

[0198] The second acquisition module is used to acquire multiple original road feature data, multiple vehicle dynamic parameters, and vehicle travel distance. Among them, the multiple original road feature data refers to road data without location compensation, which is used to represent the initial geometric structure and initial shape features of the road, and the vehicle travel distance refers to the distance traveled by the vehicle within a preset unit time.

[0199] The compensation module is used to perform position compensation on multiple original road feature data based on the vehicle's travel distance, thereby obtaining multiple road feature data.

[0200] In one possible design, the second computing module 703 includes:

[0201] The first calculation unit is used to calculate the radius of curvature of a curve based on multiple road feature data; wherein, the radius of curvature of a curve is used to represent the degree of curvature of a curve in the road.

[0202] The second calculation unit is used to calculate the safe vehicle speed based on the curve radius, road surface adhesion coefficient, and preset gravitational acceleration.

[0203] In one possible design, the trigger module 704 includes:

[0204] The first triggering unit is used to trigger the first-level warning strategy in response to the difference being less than a preset first threshold; wherein, the first-level warning strategy refers to generating and playing a voice prompt to remind the driver to slow down.

[0205] The second triggering unit is used to trigger a second-level warning strategy in response to a difference greater than a first threshold and less than a preset second threshold; wherein, the first threshold is less than the second threshold, and the second-level warning strategy refers to causing the warning icon in the vehicle's dashboard to flash to remind the driver to slow down.

[0206] The vehicle speed warning device provided in this embodiment can perform... Figure 2 and Figure 3 The technical solution of the vehicle speed warning method embodiment shown herein, its implementation principle and technical effect are similar to Figure 2 and Figure 3 The embodiment of the vehicle speed warning method shown is similar and will not be described in detail here.

[0207] Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 80 includes at least one processor 810 and a memory 820. The electronic device also includes a communication component 830. The processor 810, memory 820, and communication component 830 are connected via a bus 840.

[0208] In a specific implementation, at least one processor 810 executes computer execution instructions stored in memory 820, causing at least one processor 810 to implement a vehicle speed warning method according to the above embodiment.

[0209] The specific implementation process of processor 810 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0210] In the above embodiments, it should be understood that the processor 810 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0211] The memory 820 may include high-speed RAM memory, and may also include non-volatile memory NVM, such as at least one disk storage.

[0212] Bus 840 can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Bus 840 can be divided into address bus, data bus, control bus, etc. For ease of illustration, the bus 840 in the accompanying drawings of this application is not limited to only one bus or one type of bus.

[0213] The above description of the functions implemented by electronic devices and main control devices has introduced the solutions provided by the embodiments of the present invention. It is understood that, in order to implement the above functions, the electronic device or main control device includes hardware structures and / or software modules corresponding to the execution of each function. By combining the units and algorithm steps of the various examples described in the embodiments of the present invention, the embodiments of the present invention can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the technical solutions of the embodiments of the present invention.

[0214] This application also provides a computer-readable storage medium storing computer-executable instructions. When executed by a processor, these instructions are used to implement a vehicle speed warning method according to the above embodiments. In the specific implementation of the aforementioned vehicle speed warning method, each module can be implemented as a processor.

[0215] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0216] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in application-specific integrated circuits (ASICs). Alternatively, the processor and the readable storage medium can exist as discrete components in an electronic device or a host device.

[0217] This application also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement a vehicle speed warning method according to the above embodiments.

[0218] The computer program is stored in a readable storage medium, and at least one processor can read the computer program from the readable storage medium and execute the computer program to perform the scheme provided in any of the above embodiments.

[0219] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disk, or optical disk.

[0220] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for early warning of vehicle speed, characterized in that, include: Acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle; The road surface adhesion coefficient is calculated based on the multiple vehicle dynamic parameters and a preset hidden Markov model; wherein the road surface adhesion coefficient is used to represent the frictional force between the vehicle's tires and the road surface. Based on the multiple road feature data and the road surface adhesion coefficient, a safe vehicle speed is calculated; wherein, the safe vehicle speed refers to the maximum permissible speed at which the vehicle can safely pass through the road, and the safe vehicle speed is used to compare with the real-time speed of the vehicle to determine whether the vehicle's driving state is safe or unsafe, and the multiple vehicle dynamic parameters include the real-time speed; A tiered warning strategy is triggered based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the vehicle speed according to the severity of the speeding.

2. The vehicle speed warning method according to claim 1, characterized in that, The hidden Markov model includes a state transition matrix and a confusion matrix. The calculation of the road surface adhesion coefficient based on the multiple vehicle dynamic parameters and the preset hidden Markov model includes: Based on the multiple vehicle dynamic parameters, the state transition matrix and the confusion matrix are constructed; wherein, the state transition matrix is ​​used to represent the probability of mutual transformation between multiple preset road surface categories, the multiple road surface categories including dry cement road surface, wet and slippery cement road surface, icy and snowy road surface and dirt road surface, and the confusion matrix is ​​used to establish the probability mapping relationship between the multiple road surface categories and the multiple vehicle dynamic parameters; Based on the state transition matrix, the confusion matrix, and the multiple vehicle dynamic parameters, state identification is performed to obtain the target road surface category corresponding to the road. The road surface adhesion coefficient is determined based on the target road surface category corresponding to the road.

3. The vehicle speed warning method according to claim 2, characterized in that, The step of performing state identification based on the state transition matrix, the confusion matrix, and the multiple vehicle dynamic parameters to obtain the target road surface category corresponding to the road includes: Obtain the vehicle model; The state transition matrix and the confusion matrix are adjusted according to the vehicle model to obtain the adjusted state transition matrix and the adjusted confusion matrix; Based on the adjusted state transition matrix, the adjusted confusion matrix, and the multiple vehicle dynamic parameters, state identification is performed to obtain the target road surface category corresponding to the road.

4. The vehicle speed warning method according to claim 1, characterized in that, Before acquiring multiple road feature data and multiple vehicle dynamic parameters, the process also includes: The system acquires multiple raw road feature data, multiple vehicle dynamic parameters, and vehicle travel distance. The raw road feature data refers to road data without location compensation, which is used to represent the initial geometric structure and initial shape features of the road. The vehicle travel distance refers to the distance traveled by the vehicle within a preset unit of time. The multiple original road feature data are obtained by performing position compensation based on the vehicle's travel distance.

5. The vehicle speed warning method according to claim 1, characterized in that, The step of calculating the safe vehicle speed based on the multiple road feature data and the road surface adhesion coefficient includes: Based on the multiple road feature data, the radius of curvature of the curve is calculated; wherein, the radius of curvature of the curve is used to represent the degree of curvature of the curve in the road; The safe vehicle speed is calculated based on the radius of curvature of the curve, the road surface adhesion coefficient, and the preset gravitational acceleration.

6. The vehicle speed warning method according to claim 1, characterized in that, The step of triggering a tiered early warning strategy based on the numerical range of the difference between the real-time speed and the safe vehicle speed includes: In response to the difference being less than a preset first threshold, a first-level warning strategy is triggered; wherein, the first-level warning strategy refers to generating and playing a voice prompt to remind the driver to slow down; In response to the difference being greater than the first threshold and less than a preset second threshold, a second-level warning strategy is triggered; wherein the first threshold is less than the second threshold, and the second-level warning strategy refers to causing the warning icon in the vehicle's dashboard to flash to remind the driver to slow down.

7. A vehicle speed warning device, characterized in that, include: The first acquisition module is used to acquire multiple road feature data and multiple vehicle dynamic parameters; wherein, the multiple road feature data are used to represent the structure and shape of the road, and the multiple vehicle dynamic parameters are used to represent the driving state of the vehicle; The first calculation module is used to calculate the road surface adhesion coefficient based on the multiple vehicle dynamic parameters and a preset hidden Markov model; wherein the road surface adhesion coefficient is used to represent the friction force between the vehicle's tires and the road surface. The second calculation module is used to calculate a safe vehicle speed based on the multiple road feature data and the road surface adhesion coefficient; wherein, the safe vehicle speed refers to the maximum permissible speed at which the vehicle can safely pass through the road, and the safe vehicle speed is used to compare with the real-time speed of the vehicle to determine whether the vehicle's driving state is safe or unsafe, and the multiple vehicle dynamic parameters include the real-time speed; The triggering module is used to trigger a tiered warning strategy based on the numerical range of the difference between the real-time speed and the safe speed; wherein, the tiered warning strategy is used to remind the driver to adjust the vehicle speed according to the severity of the vehicle speeding.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; When the processor executes the computer execution instructions stored in the memory, it is used to implement the vehicle speed warning method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the vehicle speed warning method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The system includes a computer program, which, when executed by a processor, is used to implement the vehicle speed warning method as described in any one of claims 1 to 6.

Images