Emergency braking control method and device based on braking distance and driving distance estimation of self-vehicle emergency braking system and medium

Abstract

The invention provides an emergency braking control method and device based on braking distance and driving distance estimation of a self-vehicle emergency braking system and a medium, and belongs to the technical field of intelligent driving of commercial vehicles. A front vehicle multi-motion model is established, the driving distance of a front vehicle is predicted by adopting a minimum variance weighted fusion method, and the model weight is continuously optimized through a sliding window; and performing collision risk assessment based on the braking distance of the vehicle and the fusion prediction result, and executing early warning or braking according to the risk level. According to the method, the collision risk is judged by calculating the braking distance of the emergency braking system of the vehicle and estimating the driving distance of the front vehicle and using the relation among the relative distance, the braking distance and the driving distance of the front vehicle, the evaluation index TTC is abandoned, the calibration process is simplified, the suitability of the system on different vehicles and different vehicle types is improved, and the system reliability is improved. And the adaptability and robustness of the system in different environments are improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent driving technology for commercial vehicles, specifically relating to an emergency braking control method, device, and medium based on the estimation of braking distance and driving distance of the vehicle's emergency braking system. Background Technology

[0002] Currently, heavy-duty commercial vehicles are subject to strict regulation. The automatic emergency braking systems (ATCs) of these vehicles assess risk using Time to Collision (TTC), which only considers the vehicle's kinematic parameters and not its actual braking performance. Furthermore, the braking performance varies significantly between different vehicle models, making it difficult to reuse calibrated TTC thresholds and increasing calibration costs. Influenced by road conditions and environmental factors, the calibrated TTC thresholds may also vary for the same vehicle. This results in the same system, when applied to different configurations of heavy-duty commercial vehicles, either issuing warnings too early, disrupting normal driving, or issuing warnings too late, failing to effectively avoid collision risks.

[0003] In the multi-model fusion scheme of related technologies, the weight allocation adopts equal weight or empirical weight, without taking into account the difference in prediction accuracy of each model in different scenarios. As a result, the model with higher prediction accuracy cannot play its full role, while the model with lower prediction accuracy drags down the overall fusion effect. The stability and accuracy of the fusion result are insufficient, making it difficult to cope with complex and ever-changing road environments.

[0004] The parameters and error data of the relevant technology prediction models rely on a fixed historical sample set and no dynamic update mechanism has been established. As the type of road changes and the driving habits of the vehicles in front change, the reference value of the historical samples gradually decreases, the adaptability of the model parameters to the actual scenario continues to decline, and the prediction accuracy and risk assessment accuracy will be significantly reduced after later use. Summary of the Invention

[0005] This invention provides an emergency braking control method based on the braking distance and travel distance prediction of the vehicle's emergency braking system. The method calculates the braking distance of the vehicle's emergency braking system and predicts the travel distance of the preceding vehicle, and uses the relationship between relative distance, braking distance and the travel distance of the preceding vehicle to determine the collision risk. It abandons the TTC evaluation index, simplifies the calibration process, improves the system's adaptability to different vehicles and models, and enhances the system's adaptability and robustness in different environments.

[0006] The methods include: S101: Obtain the vehicle's status information and calculate the braking distance of the vehicle's automatic emergency braking system in the current state; S102: Obtain the running information of the vehicle in front, model the motion state of the vehicle in front during the braking time of the vehicle itself, and obtain multiple travel distance prediction results of the vehicle in front during the braking time. S103: Based on the historical error data of the preceding vehicle's driving distance prediction, the minimum variance method is used to weight and fuse the multiple preceding vehicle driving distance prediction results obtained in S102. S104: Configure the constraint that the sum of the weights of each prediction result is 1, and solve for the optimal weights of each predicted distance of the preceding vehicle corresponding to the weighted fusion in S103. S105: Quantify the uncertainty of the prediction error of the preceding vehicle's driving distance after weighted fusion in S103, derive the probability distribution of the prediction error and determine its uncertainty interval; S106: The historical error data of the predicted travel distance of the preceding vehicle is continuously updated using the sliding window method. Based on the updated historical error data, the optimal weight in S104 and the uncertainty interval in S105 are recalculated. S107: Based on the vehicle braking distance obtained in S101, the preceding vehicle travel distance updated in S106, and the uncertainty interval, calculate the collision risk between the vehicle and the preceding vehicle, and perform the corresponding braking operation according to the collision risk result.

[0007] According to another embodiment of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the emergency braking control method based on the estimation of braking distance and driving distance of a vehicle emergency braking system.

[0008] According to another embodiment of this application, a storage medium is also provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the emergency braking control method based on the estimation of braking distance and driving distance of the vehicle emergency braking system are implemented.

[0009] As can be seen from the above technical solutions, the present invention has the following advantages: The method provided by this invention is based on the three-stage physical characteristics of the braking process, calculated using self-vehicle braking distance. Combined with experimental parameters and a specified formula derivation, the output braking distance accurately matches the actual braking conditions of heavy commercial vehicles, reflecting the real braking space requirements under current initial speed, load, and other conditions. The method for predicting the distance to the preceding vehicle constructs three types of models: uniform speed, uniform acceleration, and nonlinear motion, covering common motion patterns of preceding vehicles in road scenarios. These models are then weighted and fused using the minimum variance method, allowing the prediction results to fully integrate the adaptability advantages of each model and reduce prediction bias in specific scenarios.

[0010] The optimal weight solution in this invention is achieved by setting a constraint that the sum of the weights is 1, combined with the derivation of the Lagrange function. This ensures that the weight allocation conforms to mathematical logic, minimizes the fusion error, keeps the fluctuation of the fusion result within a controllable range, and improves the stability of the prediction.

[0011] The uncertainty quantification of prediction error in this invention employs a normal distribution description and a 3σ interval definition, clearly defining the fluctuation boundaries of the predicted value and providing an error reference dimension for collision risk assessment. The sliding window method dynamically updates historical error data and adjusts the optimal weights and uncertainty intervals in real time, ensuring that model parameters closely follow changes in driving scenarios and maintaining stable prediction accuracy and scenario adaptability over long-term use.

[0012] The collision risk calculation of this invention integrates the vehicle's braking distance, the updated preceding vehicle's travel distance, and the uncertainty range to achieve a refined classification of risk levels. It can ensure braking redundancy in extreme scenarios and avoid unnecessary driving intervention in normal scenarios. Attached Figure Description

[0013] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 Flowchart of an embodiment of an automatic emergency braking control method for commercial vehicles; Figure 2 A schematic diagram illustrating the braking distance calculation for a vehicle's emergency braking system; Figure 3 A diagram illustrating the current collision risk of a vehicle; Figure 4 Flowchart of automatic emergency braking control method for commercial vehicles; Figure 5 This is a schematic diagram of an electronic device. Detailed Implementation

[0015] like Figure 1 As shown, the emergency braking control method based on the braking distance and travel distance prediction of the vehicle's emergency braking system provided by the present invention calculates the braking distance of the vehicle under the current vehicle state and predicts the travel distance of the preceding vehicle within the braking time of the vehicle in real time. By comparing the relationship between the braking distance, travel distance and relative distance, the method makes a decision for the vehicle.

[0016] The control method involved in this invention calculates the braking distance of the vehicle's emergency braking system and estimates the travel distance of the vehicle in front. It uses the relationship between relative distance, braking distance and travel distance of the vehicle in front to determine the collision risk. It abandons the TTC evaluation index, simplifies the calibration process, reduces calibration costs, improves the system's adaptability to different vehicles and models, and improves the system's adaptability and robustness in different environments.

[0017] The control method of this invention is mainly divided into three parts. The first part is the calculation of the braking distance of the vehicle's emergency braking system. This part mainly calculates the braking distance of the vehicle in the current state in real time through the vehicle's sensor information and the braking curve of the automatic emergency braking system. Figure 2 The second part is the prediction of the distance to the vehicle ahead. This part mainly involves kinematic modeling of the vehicle ahead, using the minimum variance method to weightedly fuse the kinematic models, solving for the optimal weights of each model using the Lagrangian function, quantifying the uncertainty of the error, and continuously updating the prediction error and optimal weights using the sliding window method to achieve model adaptation. The third part is the collision risk calculation, which involves the relationship between the calculated braking distance and the predicted distance to the vehicle ahead, such as... Figure 3 The aim is to assess the current collision risk of the vehicle and ultimately develop a simple and adaptable control method for an automatic emergency braking system for heavy-duty commercial vehicles.

[0018] The following describes in detail the emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system, as per this application. Specific details, such as particular system structures and technologies, are presented for illustrative purposes and not for limitation, to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details.

[0019] It should be understood that, when used in this specification, the term "comprising" indicates the presence of the described feature, integral, step, operation, element, and / or component, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0020] The terms "one embodiment" or "some embodiments" used in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this application do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 4 The diagram shows a flowchart of an emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system, according to a specific embodiment. The method includes: S101: Obtain the vehicle's status information and calculate the braking distance of the vehicle's automatic emergency braking system in the current state.

[0023] S101 specifically includes the following steps: S1011: The initial speed of the vehicle is currently being collected. And retrieve the corresponding deceleration response time of the braking system. deceleration rise time , and the stable deceleration a of the braking system; In some embodiments, the initial speed of the vehicle under its current operating state is collected in real time via a vehicle body sensing module. The deceleration response time of the braking system, matched to the current vehicle model, is retrieved from a pre-stored test data table of the braking system. .time The value range is 0.1~0.2s. Deceleration rise time. ,time The value range is 0.4~0.6s, and the stable deceleration a that the braking system can achieve is also retrieved.

[0024] S1012: Based on the characteristic that the deceleration is 0 during the deceleration response stage, combined with the collected data... and Calculate the travel distance during the deceleration response phase. ; In some embodiments, during the deceleration response phase, the braking system has not yet generated deceleration, and the vehicle maintains its initial speed. Traveling at a constant speed, the travel time is the value retrieved. Through calculation and The product of these two numbers gives the distance traveled during that stage. .

[0025] S1013: Approximate the deceleration rise phase as a linear rise process, utilizing... Formula (1); Combination a, Calculate the driving distance ; In some embodiments, during the deceleration rise phase, the deceleration increases linearly from 0 to a steady deceleration α, and the average deceleration during the deceleration response phase is... The travel time is ,Will a, Substituting into formula (1), the travel distance s2 for this stage is calculated.

[0026] S1014: Based on the characteristic that the average deceleration during the deceleration steady-state phase is 'a', utilizing... Formula (2), combined with a, Calculate the travel distance during this stage. ; In some embodiments, during the deceleration stabilization phase, the braking system maintains a stable deceleration rate 'a', a, Substitution The travel distance for this stage was calculated. .

[0027] S1015: Combination , , ,use Formula (3) is used to calculate the total braking distance s of the vehicle's automatic emergency braking system in the current state.

[0028] In some embodiments, the deceleration response phase The deceleration acceleration phase During the deceleration stabilization phase Substituting into formula (3) and performing a combined calculation, we obtain the total braking distance s required for the automatic emergency braking system of the vehicle to complete braking in the current state. In this way, during the deceleration stabilization phase, the deceleration of the braking system remains at a stable value a, and the vehicle continues to decelerate at this deceleration until it stops. Substituting into formula (2) will yield the travel distance during this phase.

[0029] The complete braking process of a vehicle consists of three stages: response, acceleration, and stabilization. The total braking distance is the sum of the distances traveled in these three stages. Substituting these distances into formula (3) allows us to combine the distances of each stage into the final total braking distance. By integrating the calculation results of each stage, we obtain distance parameters that reflect the complete braking process of the vehicle.

[0030] S102: Obtain the running information of the vehicle in front, model the motion state of the vehicle in front during the braking time of the vehicle itself, and obtain multiple travel distance prediction results of the vehicle in front during the braking time.

[0031] S102 specifically includes the following steps: S1021: Collect the current operating parameters of the vehicle in front and retrieve the total braking time t of the vehicle itself; In some embodiments, the current speed of the vehicle in front can be obtained in real time through a sensing module configured in the vehicle. Current acceleration ; retrieve the total braking time t of the vehicle determined in S101, where t is the deceleration response time. deceleration rise time The sum of the deceleration settling time and the total time.

[0032] S1022: Based on the setting of the vehicle in front moving at a constant speed, the predicted travel distance of the vehicle in front under this state is calculated by combining the collected parameters; In some embodiments, it is defined that the preceding vehicle maintains its current speed during the braking time t of the following vehicle. Uniform motion, Multiplying this directly by t yields the predicted distance traveled by the vehicle ahead under this motion state. .

[0033] S1023: Based on the uniform acceleration motion of the preceding vehicle, based on Formula (4) is used to calculate the predicted travel distance of the vehicle in front under this condition; In some embodiments, the preceding vehicle maintains its current acceleration during the braking time t of the preceding vehicle. Performing uniformly accelerated or decelerated motion, , Substitute t The predicted travel distance of the vehicle in front under motion conditions is calculated. .

[0034] S1024: Based on the nonlinear motion of the preceding vehicle, through... Formula (5) is used to construct an acceleration change model, and combined with Formula (6) is used to calculate the predicted travel distance of the vehicle in front under this condition; In the formula, For time variables, This is the acceleration change curve over future time. The acceleration of the target vehicle ahead is obtained by fitting historical data, and T is a time constant identified by historical data.

[0035] In some embodiments, the preceding vehicle is configured to undergo nonlinear acceleration and deceleration motion during the braking time t of the following vehicle. Formula (5) constructs a first-order inertial model of acceleration change, where T is obtained by fitting historical operating data of the preceding vehicle, and is determined by identifying historical data of the preceding vehicle. The model is integrated, combined with... Formula (6) is used to calculate the predicted travel distance of the vehicle in front under this motion state. .

[0036] S1025: Summarize the results obtained from S1022, S1023, and S1024 to form multiple predicted travel distances of the preceding vehicle during the braking time of the following vehicle.

[0037] In some embodiments, the uniform motion prediction distance obtained in S1022 is... The predicted distance of uniformly accelerated motion obtained from S1023 The nonlinear motion prediction distance obtained from S1024 The data is organized into a set and used as input data for weighted fusion.

[0038] In this embodiment, the vehicle in front may experience nonlinear acceleration and deceleration during actual driving, using a first-order inertial model. It can describe the continuous change process of its acceleration, based on Integrating the model yields the travel distance under nonlinear motion. This covers complex scenarios involving nonlinear changes in the acceleration and deceleration of the vehicle in front, making the prediction results more consistent with the diverse motion states of the vehicle in front on real roads.

[0039] This embodiment combines the prediction results of different motion models corresponding to different driving states of the vehicle in front, forming a prediction dataset covering multiple scenarios. This allows the fusion result to integrate prediction information from different scenarios.

[0040] S103: Based on the historical error data of the preceding vehicle's travel distance prediction, the minimum variance method is used to perform weighted fusion of multiple preceding vehicle travel distance prediction results obtained in S102.

[0041] S103 specifically includes the following steps: S1031: Retrieve historical sample data of the predicted travel distance of the vehicle ahead, including historical prediction values ​​of each model. i is the model index, k is the sample index, and k is the true distance for each sample. ; In some embodiments, the most recent N sets of sample data for the predicted travel distance of the preceding vehicle are retrieved. Each set of data contains historical prediction values ​​corresponding to the motion model in S102. And the true distance corresponding to this set of samples. Historical sample data records the predictions made by each model in the past, while the actual distance is the result of actual driving. Only by retrieving this data can we compare how much the model's predicted value differs from the actual value.

[0042] The motion models include uniform motion models, uniformly accelerated motion models, and nonlinear motion models.

[0043] It is The distance traveled by the vehicle at the sample time is calculated using formula (8). , No. Secondary relative distance and the Secondary relative distance By plugging the formula (8) into it, we can get the result.

[0044] S1032: Let the first The predicted values ​​of each model are , , No. The weights of each model are The total predicted value after fusion is , No. The historical prediction error of each model is Formula (7) , For historical sample size, For the first The next true distance, where: Formula (8) In the formula, This represents the distance traveled by the vehicle. For the first Secondary relative distance For the first Secondary relative distance; Combining formulas (7) and (8), calculate the historical prediction error for each model. .

[0045] In some embodiments, for each model, the historical predictions of that model are... and the corresponding actual distance Substituting into formula (7), we can calculate the historical prediction error of this model for the kth sample. By calculating the predictions for all N samples of each model, we can obtain the historical prediction error data for each model.

[0046] S1033: Predictions based on current models Distance from reality ,pass Formula (10) determines the corrected error for each model. ; In some embodiments, for each model prediction value output at the current time S102 Combined with the actual distance at the current moment ,Bundle and Substituting into formula (10), the corrected error for each model can be calculated. The corrected error is the direct deviation between the model's predicted value and the actual distance at the current moment. It can be calculated using formula (10) and directly reflects the model's prediction performance in the current scenario.

[0047] S1034: Based on the corrected errors of each model, through Formula (11) constructs the variance expression for the fused error; In some embodiments, based on the corrected error of each model There is also the covariance of errors between different models. Expanding the variance of the fused error according to formula (11) yields the result containing the weights of each model. The variance expression. This embodiment can also be converted into matrix form: The matrix form of formula (12) is convenient for later calculations.

[0048] S1035: With the objective of minimizing the variance of the fused error, assign corresponding weights to the multiple preceding vehicle distance prediction results obtained in S102. The weighted fusion is then performed to obtain the total predicted value.

[0049] In some embodiments, minimizing the variance of the fused error is the objective, and corresponding weights are assigned to each predicted distance traveled by the preceding vehicle obtained in S102. The weighted fusion of each prediction result and its weight is used to construct the total predicted distance of the vehicle in front. This embodiment constructs the total predicted value by combining the prediction results of multiple models, ensuring the accuracy of the predicted distance of the vehicle in front.

[0050] S104: Configure the constraint that the sum of the weights of each prediction result is 1, and solve for the optimal weights of each preceding vehicle's travel distance prediction result corresponding to the weighted fusion in S103.

[0051] S104 specifically includes the following steps: S1041: Define the weight vector for each preceding vehicle's travel distance prediction model, and configure the constraint condition that the sum of each weight is 1. The constraint is explicitly stated in the form of formula (13); In some embodiments, the weights of each preceding vehicle distance prediction model in S103 are integrated into a weight vector ω. The elements of ω are the weights of each model. i .

[0052] This embodiment explicitly states the constraint: the sum of the weights of all models equals 1, and this constraint is expressed by formula (13). ω i =1, which can also be written as the transpose of the all-1 vector l and the weight vector ω equals 1.

[0053] S1042: Combining the matrix expression of the fused error variance in S103, construct the Lagrange function: Formula (14); where λ is the Lagrange multiplier; In some embodiments, the matrix form of formula (12) for the fused error variance in S103 is retrieved, where Σ is the covariance matrix of the errors of each model. The Lagrange multiplier λ is configured to transform the constraints with a weight sum of 1 into terms of lTw-1, and the Lagrange function is constructed according to formula (14). Here, instead of dealing with the constraints separately, we can directly satisfy both conditions by finding the extrema of the function, simplifying the optimization process.

[0054] S1043: Take the partial derivatives of the Lagrange function with respect to the weight vector and the Lagrange multiplier respectively, and set the partial derivatives to 0 to obtain... Formula (15) Formula (16) In some embodiments, the partial derivative of the Lagrange function with respect to the weight vector ω is obtained according to the matrix differentiation rules. The partial derivative is 2Σw. The partial derivative is λl. Setting the partial derivative to 0, we obtain formula (15). Taking the partial derivative of the Lagrange function with respect to λ, the result is... Set it to 0 to get Formula (16).

[0055] Here, the optimization problem is transformed into solving a system of linear equations, making the steps clearer and more feasible.

[0056] S1044: Solve for the expression of the weight vector from the partial derivative equations, substitute it into the constraints, and solve for the Lagrange multipliers to obtain: Formula (17) Formula (18) Formula (19) Finally, the optimal weight vector is obtained. Formula (20).

[0057] In some embodiments, the expression for solving the weight vector w from formula (15) is: Substituting this expression into formula (16) Thus, formula (18) is obtained. After simplification, λ is solved as follows: That is, formula (19). Here, the bivariate system of equations is broken down into step-by-step solutions, which reduces the computational complexity and allows for accurate determination of the Lagrange multipliers.

[0058] S1045: Substitute the obtained Lagrange multipliers into the weight vector expression to obtain the optimal weight vector. If the motion models are independent... Then the weight matrix simplifies to: Formula (21) In some embodiments, the λ obtained from formula (19) is substituted into the w expression of formula (17) to obtain the optimal weight vector. Formula (20). If the prediction errors of each model are independent and the covariance matrix Σ is a diagonal matrix, then the optimal weights simplify to... Formula (21) This is the variance of the error of the i-th model. This allows us to obtain the optimal weights to improve prediction accuracy. Furthermore, the simplified calculations when the errors are independent can reduce computational load and speed up the solution process.

[0059] S105: Quantify the uncertainty of the prediction error of the preceding vehicle's driving distance after weighted fusion in S103, derive the probability distribution of the prediction error, and determine its uncertainty interval.

[0060] In some embodiments, the fluctuation of prediction error is random. The normal distribution can effectively describe the statistical regularity of such independent small perturbations. The 3σ principle can cover most error situations, ensuring that the uncertainty interval can reflect the error range in most real-world scenarios.

[0061] S106: The historical error data of the predicted travel distance of the preceding vehicle is continuously updated using the sliding window method. Based on the updated historical error data, the optimal weight in S104 and the uncertainty interval in S105 are recalculated.

[0062] In some embodiments, the motion characteristics of the preceding vehicle may change dynamically with the driving scenario. The timeliness of historical error data directly affects the adaptability of the weights and uncertainty intervals. The sliding window ensures that the data used in the calculation always reflects recent error characteristics by incorporating new data and removing old data. This allows the optimal weights and uncertainty intervals to adapt to changes in the motion state of the preceding vehicle in real time, enabling the model to adapt online and improving the overall system's adaptability to dynamic driving scenarios.

[0063] S107: Based on the vehicle braking distance obtained in S101, the preceding vehicle travel distance updated in S106, and the uncertainty interval, calculate the collision risk between the vehicle and the preceding vehicle, and perform the corresponding braking operation according to the collision risk result.

[0064] In some embodiments, collision risk is determined by the matching relationship between the space required for the vehicle's braking, the relative distance between the two vehicles, and the distance traveled by the vehicle in front. Combining this with the lower limit of the uncertainty range can cover extreme risk scenarios where the distance traveled by the vehicle in front is the shortest. Layered threshold judgments can achieve risk control. This enables dynamic and accurate assessment of collision risk, preventing oversensitivity from affecting normal driving and avoiding overlooking high-risk scenarios due to delayed reactions, thus ensuring driving safety.

[0065] In one embodiment of the present invention, based on step S105, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S105 specifically includes the following steps: S1051: Analyze the characteristics of the errors after correction for each motion model. Based on the central limit theorem, assume that the errors of the corrected motion models follow a normal distribution. Let the errors of the corrected motion models be... Formula (22); In some embodiments, the corrected errors of each motion model are analyzed, which is the error in formula (10). Due to the characteristics of these errors, most of them come from independent small disturbances in the driving scenario, which meets the application premise of the central limit theorem.

[0066] In this embodiment, the corrected error for each motion model is defined, following a mean of 0 and a variance of . The normal distribution, this assumption is used ,in It is the variance of the error after the i-th motion model is corrected.

[0067] S1052: Retrieve the optimal weight vector obtained from S104, combine it with the corrected errors of each motion model, and calculate the fused prediction error using formula (23). Formula (23).

[0068] In some embodiments, the optimal weights of each motion model calculated by S104 are retrieved. The corrected error of each motion model The product is multiplied by the corresponding optimal weight, and then all products are summed to obtain the fused prediction error. Here, the fused error is designed to reflect the error characteristics of each motion model, without ignoring the prediction bias of any single model.

[0069] S1053: Substituting the optimal weight vector into the variance expression of the fusion error, the minimum variance of the fusion error is derived, corresponding to... Formula (24); In some embodiments, the optimal weight vector obtained in S104 is substituted into the formula for calculating the variance of the fusion error, and then combined with the covariance matrix of the errors of each motion model, the minimum variance of the fused error is derived. In formula (24), l is a vector of all 1s, and Σ is the covariance matrix of the motion model error. The variance obtained in this way is the most stable error fluctuation after fusion, and there will be no excessive variance due to unreasonable weights.

[0070] S1054: Based on the 3σ principle of normal distribution and combined with the minimum variance result, the uncertainty interval of the prediction error after fusion is determined by formula (25); Formula (25) In some embodiments, according to the 3σ principle of normal distribution, that is, the normal distribution probability will fall within the range of mean ± 3 times standard deviation, the uncertainty interval of the prediction error after fusion is calculated by combining the minimum variance calculated by formula (24).

[0071] This interval is defined in formula (25). This is the fused predicted distance traveled by the vehicle ahead. The defined uncertainty range is reliable enough to cover error scenarios in most real-world situations.

[0072] S1055: Detect the number of historical error samples to verify the rationality of the normal distribution assumption and ensure the validity of the quantification results.

[0073] In some embodiments, historical error sample data is retrieved to check if the sample size is large enough to meet the large sample requirement of the Central Limit Theorem. It is then checked whether each error is an independent perturbation. These two checks verify the rationality of the normal distribution assumption in S1051, ensuring that the subsequent uncertainty quantification results closely match the actual scenario. Thus, the Central Limit Theorem and the normal distribution assumption require prerequisites such as sample size and perturbation independence; checking these conditions ensures the assumptions hold.

[0074] In one embodiment of the present invention, based on step S106, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S106 specifically includes the following steps: S1061: Set the fixed size N of the sliding window and define the storage range of historical error data within the window; In some embodiments, based on the dynamic changes in the driving status of the preceding vehicle and the system's computing power, a fixed size of N is set for the sliding window, where N is a positive integer and its value must cover a sufficient number of historical samples to ensure the validity of the error statistics. The window explicitly stores only the prediction error data of the preceding vehicle's driving distance from the most recent N frames, thus defining the effective range of the historical error data.

[0075] S1062: Collects real-time operating data of the vehicle in front through the vehicle's sensing module, combined with... Calculate the actual distance traveled by the vehicle in front at the current moment, and then based on... , Obtain the prediction errors of each current model; In some embodiments, the current speed, current acceleration, and relative distance d between the vehicle and the preceding vehicle are collected in real time using sensing devices such as millimeter-wave radar and vision sensors configured in the vehicle. Based on formula (8) and the collected travel distance of the vehicle, the actual distance traveled by the preceding vehicle at the current moment is calculated. The historical prediction error εi,k of each current model is calculated using formula (7), and the corrected error of each current model is calculated using formula (10). .

[0076] S1063: Add the prediction errors of each model to the sliding window, delete the historical error data of the earliest frame in the window, and maintain the total amount of data in the window at N; In some embodiments, the current prediction errors of each model obtained in S1062 are used as new data frames and added to a preset sliding window. The oldest historical error data frame stored in the window is retrieved and removed from the window to ensure that N consecutive frames of error data are always maintained within the sliding window.

[0077] S1064: Based on the historical error data within the updated window, recalculate the variance of the errors for each motion model. and the covariance of errors between different motion models Construct the updated error covariance matrix; In some embodiments, based on the updated N frames of error data within the window, the variance of the error for each motion model is recalculated. This is the average of the squared errors. For any two different motion models, recalculate the covariance of their errors. That is, the average of the product of the errors of the two models. All variances... With covariance Arranged according to their corresponding positions, the updated motion model error covariance matrix is ​​constructed. This yields an error covariance matrix that matches the current driving state, eliminating weight calculation biases caused by changes in error characteristics and providing a reliable basis for dynamic adjustment of optimal weights.

[0078] S1065: Substitute the updated covariance matrix into... Resolve for the optimal weights, and combine , Recalculate the uncertainty interval.

[0079] In some embodiments, if the errors of the various motion models are not perfectly correlated, the updated covariance matrix is ​​substituted into... Then, recalculate the optimal weight vector w* for each model. If the motion model errors are independent, directly use... Simplify the calculation of optimal weights; then substitute the updated covariance matrix into... The minimum variance of the fusion error was re-derived, combined with The 3σ principle is used to redefine the uncertainty range of the prediction error after fusion. This ensures that the optimal weights continuously align with the current prediction error characteristics, making the uncertainty range more consistent with the actual error fluctuation range, thus improving the real-time performance and accuracy of subsequent collision risk calculations.

[0080] In one embodiment of the present invention, based on step S107, the following is a possible embodiment and its specific implementation will be described in a non-limiting manner. S107 specifically includes the following steps: S1071: Retrieve total braking distance S and total predicted value of the preceding vehicle. And the uncertainty range, to obtain the driver's reaction time With braking margin ; In some embodiments, the total braking distance S calculated in S101 is retrieved, and the total predicted value of the preceding vehicle updated in S106 is retrieved. And the uncertainty range. The driver reaction time, matched to the current vehicle model, is obtained from the pre-stored test calibration database of the braking system. Braking margin to ensure braking safety .

[0081] S1072: According to Formula (26) clarifies the conditions for avoiding collisions between the vehicle and the vehicle in front; In some embodiments, with the aim of avoiding a collision between the vehicle and the vehicle in front, the determination is based on the following criteria: Formula (26), that is, the relative distance l between the current vehicle and the vehicle in front must satisfy... , where D is the predicted distance traveled by the preceding vehicle after the S106 update. This condition directly reflects whether a collision can be avoided.

[0082] S1073: Calibrate the risk assessment benchmark by combining the lower limit of the uncertainty interval, based on... Formula (27) determines the collision warning threshold; In some embodiments, considering that the lower limit of the uncertainty interval corresponds to the extreme risk scenario where the distance traveled by the preceding vehicle is minimized, the determined uncertainty interval is:

[0083] lower limit value As a calibration benchmark for risk assessment; based on formula (27), combined with The collision warning threshold is calculated based on the vehicle's initial speed v0 and the calibrated baseline. This threshold serves as the critical criterion for initiating the warning operation. Here, the risk assessment baseline is aligned with extreme risk scenarios, while the warning threshold takes into account the driver's reaction characteristics, avoiding warning delays or misjudgments due to failure to consider extreme situations.

[0084] S1074: Collect the real-time relative distance l between the current vehicle and the vehicle in front, compare the core conditions, warning thresholds and braking judgment conditions, and classify the collision risk level; In some embodiments, the relative distance l between the current vehicle and the vehicle in front is collected in real time by the vehicle sensing module, and l is compared with... The main conditions of formula (26) and the braking judgment conditions of the warning threshold determined by S1073 are compared. If l is greater than the warning threshold, it is judged as no collision risk. If l is between the upper limit of the braking judgment condition and the warning threshold, it is judged as low collision risk; if l meets the braking judgment condition, it is judged as high collision risk. This realizes the fine classification of collision risk, making subsequent operations more targeted.

[0085] S1075: Based on the collision risk level, perform the corresponding operation: maintain the current state if there is no risk, issue a warning if there is a low risk, or initiate automatic braking if there is a high risk.

[0086] In some embodiments, when no collision risk is determined, the vehicle maintains its current driving state and the braking system is in standby mode; when a low collision risk is determined, the vehicle issues a collision warning to the driver via an audible and visual warning device; when a high collision risk is determined, the vehicle automatically activates the emergency braking system and performs braking operations according to the deceleration curve corresponding to the braking distance determined in S101. This avoids excessive intervention affecting the driving experience, ensures active protection in high-risk scenarios, and makes the operation of the automatic emergency braking system more in line with actual usage needs.

[0087] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0088] like Figure 5 As shown, this application also provides an electronic device, including a display module 103, a memory 102, a processor 101, a communication module 104, and a computer program stored in the memory and executable on the processor 101. When the processor 101 executes the program, it implements the steps of an emergency braking control method based on the estimation of braking distance and driving distance of a vehicle's emergency braking system.

[0089] In embodiments of the present invention, electronic devices include, but are not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also refer to various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the embodiments described and / or claimed herein.

[0090] In this embodiment, processor 101 may be implemented using at least one of an application-specific integrated circuit, a programmable logic device, a field-programmable gate array, a processor, a controller, a microcontroller, a microprocessor, or an electronic unit designed to perform the functions described herein. In some cases, such an implementation may be implemented within a controller. For software implementation, implementations such as processes or functions may be implemented with separate software modules that allow the performance of at least one function or operation. Software code may be implemented by a software application (or program) written in any suitable programming language, and the software code may be stored in memory and executed by the controller.

[0091] The display module 103 is used to display information input by the user or information provided to the user. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like.

[0092] The memory 102 can be used to store software programs and various data. The memory 102 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0093] The communication module 104 transmits radio signals to and / or receives radio signals from at least one of a base station, an external terminal, and a server. Such radio signals may include voice call signals, video call signals, or various types of data sent and / or received according to text and / or multimedia messages.

[0094] The present invention also provides a storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of the emergency braking control method based on the estimation of braking distance and driving distance of the vehicle emergency braking system.

[0095] The storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0096] The storage medium stores a program product capable of implementing the methods described above in this specification. In some possible implementations, various aspects of this disclosure may also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0097] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system, characterized in that, The methods include: S101: Obtain the vehicle's status information and calculate the braking distance of the vehicle's automatic emergency braking system in the current state; S102: Obtain the running information of the vehicle in front, model the motion state of the vehicle in front during the braking time of the vehicle itself, and obtain multiple travel distance prediction results of the vehicle in front during the braking time. S103: Based on the historical error data of the preceding vehicle's driving distance prediction, the minimum variance method is used to weight and fuse the multiple preceding vehicle driving distance prediction results obtained in S102. S104: Configure the constraint that the sum of the weights of each prediction result is 1, and solve for the optimal weights of each predicted distance of the preceding vehicle corresponding to the weighted fusion in S103. S105: Quantify the uncertainty of the prediction error of the preceding vehicle's driving distance after weighted fusion in S103, derive the probability distribution of the prediction error and determine its uncertainty interval; S106: The historical error data of the predicted travel distance of the preceding vehicle is continuously updated using the sliding window method. Based on the updated historical error data, the optimal weight in S104 and the uncertainty interval in S105 are recalculated. S107: Based on the vehicle braking distance obtained in S101, the preceding vehicle travel distance updated in S106, and the uncertainty interval, calculate the collision risk between the vehicle and the preceding vehicle, and perform the corresponding braking operation according to the collision risk result.

2. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S101 specifically includes the following steps: The initial speed of the vehicle is currently being collected. And retrieve the corresponding deceleration response time of the braking system. deceleration rise time , and the stable deceleration a of the braking system; Based on the characteristic that deceleration is zero during the deceleration response phase, combined with the collected data... and Calculate the travel distance during the deceleration response phase. ; Approximating the acceleration phase as a linear upward process, using... , combined a, Calculate the driving distance ; Based on the characteristic that the average deceleration during the steady-state phase is 'a', using... , combined a, Calculate the travel distance during this stage. ; Combination , , ,use The total braking distance s of the vehicle's automatic emergency braking system in the current state is calculated.

3. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S102 specifically includes the following steps: Collect the current operating parameters of the vehicle in front and retrieve the total braking time t of the vehicle itself; Based on the setting of the vehicle in front moving at a constant speed, the predicted travel distance of the vehicle in front under this state is calculated by combining the collected parameters. Based on the uniform acceleration motion of the vehicle in front, based on Calculate the predicted travel distance of the vehicle in front under this condition; Based on the nonlinear motion of the vehicle in front, through Construct an acceleration change model and combine it with Calculate the predicted travel distance of the vehicle in front under this condition; In the formula, For time variables, This is the acceleration change curve over future time. Accelerate the target vehicle ahead; Based on the results obtained, multiple travel distance predictions are generated for the preceding vehicle during the braking time of the following vehicle.

4. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S103 specifically includes the following steps: Retrieve historical sample data for predicting the distance traveled by the vehicle in front, including historical prediction values ​​from each model. i is the model index, k is the sample index, and k is the true distance for each sample. ; Let the first The predicted values ​​of each model are , , No. The weights of each model are The total predicted value after fusion is , No. The historical prediction error of each model is , For historical sample size, For the first The next true distance, where: In the formula, This represents the distance traveled by the vehicle. For the first Secondary relative distance For the first Secondary relative distance; Calculate the historical prediction error for each model. ; Based on the predictions of the current models Distance from reality ,pass Determine the corrected error for each model ; Based on the corrected errors of each model, through Construct a variance expression for the fused error; With the goal of minimizing the variance of the fused error, the multiple predicted distances of the preceding vehicle obtained from S102 are assigned corresponding weights. The weighted fusion is then performed to obtain the total predicted value.

5. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S104 specifically includes the following steps: Define the weight vectors for each preceding vehicle's travel distance prediction model, and configure a constraint that the sum of all weights is 1. The form is clearly constrained; Based on the matrix expression of the fused error variance in S103, construct the Lagrange function: Where λ is the Lagrange multiplier; Taking the partial derivatives of the Lagrange function with respect to the weight vector and the Lagrange multiplier respectively, and setting the partial derivatives to 0, we obtain... Solve for the weight vector from the partial derivative equations, substitute it into the constraints, and solve for the Lagrange multipliers to obtain: Finally, the optimal weight vector is obtained. Substituting the obtained Lagrange multipliers into the weight vector expression yields the optimal weight vector. If the motion models are independent, the weight matrix simplifies to: 。 6. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S105 specifically includes the following steps: Analyze the characteristics of the errors after correction for each motion model. Based on the central limit theorem, assume that the errors of the corrected motion models follow a normal distribution. Let the errors of the corrected motion models be... The optimal weight vector obtained from S104 is retrieved, and the corrected errors of each motion model are combined to calculate the fused prediction error. Substituting the optimal weight vector into the variance expression of the fusion error, the minimum variance of the fusion error is derived, corresponding to... Based on the 3σ principle of normal distribution and combined with the minimum variance result, the uncertainty interval of the prediction error after fusion is determined; The number of historical error samples is checked to verify the rationality of the normal distribution assumption and ensure the validity of the quantification results.

7. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S106 specifically includes the following steps: Set a fixed size N for the sliding window and define the storage range of historical error data within the window; By collecting real-time operating data of the vehicle in front through the vehicle's sensing module, combined with... Calculate the actual distance traveled by the vehicle in front at the current moment, and then based on... , Obtain the prediction errors of each current model; The prediction errors of each model are added to the sliding window, and the historical error data of the earliest frame in the window is deleted, while maintaining the total amount of data in the window at N. Based on the historical error data within the updated window, the variance of the error for each motion model is recalculated. and the covariance of errors between different motion models Construct the updated error covariance matrix; Substitute the updated covariance matrix into Resolve for the optimal weights, and combine , Recalculate the uncertainty interval.

8. The emergency braking control method based on the estimation of braking distance and travel distance of a vehicle's emergency braking system according to claim 1, characterized in that, S107 specifically includes the following steps: Retrieve total braking distance S and total predicted value of the preceding vehicle And the uncertainty range, to obtain the driver's reaction time With braking margin ; according to Clearly define the conditions for avoiding collisions between your vehicle and the vehicle in front; The risk assessment benchmark is calibrated by combining the lower limit of the uncertainty interval, based on Determine the collision warning threshold; Collect the real-time relative distance l between the current vehicle and the vehicle in front, compare it with the core conditions, warning thresholds and braking judgment conditions, and classify the collision risk level; Depending on the collision risk level, the system will execute the corresponding actions: maintain the current state if there is no risk, issue a warning if there is a low risk, or initiate automatic braking if there is a high risk.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the emergency braking control method based on the estimation of braking distance and driving distance of the vehicle emergency braking system as described in any one of claims 1 to 8.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the emergency braking control method based on the estimation of braking distance and driving distance of the vehicle emergency braking system as described in any one of claims 1 to 8.

Images