Method and system for detecting accelerator pedal misapplication

Abstract

The present application relates to a method, system, computer storage medium, computer device and vehicle for detecting accelerator pedal misapplication. The method for detecting accelerator pedal misapplication according to one aspect of the present application comprises: obtaining vehicle driving data; processing the vehicle driving data by using a misapplication prediction model to predict whether the accelerator pedal is misapplied; judging whether the vehicle is in a predetermined acceleration scenario in response to predicting that the accelerator pedal is misapplied; and determining that the accelerator pedal is misapplied in response to judging that the vehicle is not in the predetermined acceleration scenario.

Description

Technical Field

[0001] This invention relates to the field of vehicle safety, and more specifically to a method, system, computer storage medium, computer device, and vehicle for detecting accidental pressing of the accelerator pedal. Background Technology

[0002] During vehicle driving, when faced with an emergency situation requiring emergency braking, drivers may sometimes accidentally touch the accelerator pedal or mistake it for the brake pedal due to reasons such as fatigue, lack of concentration, excessive tension, or lack of experience, leading to serious traffic accidents.

[0003] Currently, to prevent the accelerator pedal from being accidentally pressed, the pedal reaction torque can be significantly increased for a short period of time when the following distance is too close, thereby reminding the driver to avoid following too closely through touch. Alternatively, multiple cameras can be installed on the vehicle to identify the surrounding environment and automatically brake to intervene when the accelerator pedal is accidentally pressed when an obstacle is detected in front of the vehicle, in order to avoid or mitigate a collision.

[0004] However, the above methods for preventing accidental accelerator pedal depressing only utilize limited environmental information to identify such behavior. This may lead to misjudging normal driving behavior as accidental accelerator pedal depressing in real-world scenarios, resulting in intervention and severely impacting the driving experience. Summary of the Invention

[0005] To address or at least alleviate one or more of the above problems, the following technical solutions are provided.

[0006] According to a first aspect of the present invention, a method for detecting accidental accelerator pedal press is provided, comprising: acquiring vehicle driving data; processing the vehicle driving data using an accidental press prediction model to predict whether an accelerator pedal press has occurred; determining whether the vehicle is in a predetermined acceleration scenario in response to the prediction that the accelerator pedal press has occurred; and determining that the accelerator pedal press has occurred in response to the determination that the vehicle is not in the predetermined acceleration scenario.

[0007] According to an embodiment of the present invention, the method for detecting accelerator pedal misapplication includes one or more of the following vehicle driving data: vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening.

[0008] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication includes processing the vehicle driving data using a misapplication prediction model to predict whether the accelerator pedal has been misapplication, which includes: inputting the vehicle driving data into the misapplication prediction model to generate a misapplication probability value; and predicting that the accelerator pedal has been misapplication when the generated misapplication probability value is greater than a preset probability value.

[0009] According to one or more embodiments of the present invention, the method for detecting accelerator pedal misapplication includes one or more of the following predetermined acceleration scenarios: uphill acceleration, acceleration exiting a curve, acceleration during forward / reverse gear shifting, and acceleration after brake release.

[0010] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication further includes: determining the current road scene of the vehicle based on vehicle GPS data before processing the vehicle driving data using a misapplication prediction model to predict whether the accelerator pedal has been misapplication; and selecting a misapplication prediction model associated with the determined current road scene of the vehicle.

[0011] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication is trained by the following steps: extracting data features from data samples of normal vehicle driving and sample data of accelerator pedal misapplication to obtain a first training data sample set; performing negative sample augmentation on the sample data of accelerator pedal misapplication to obtain a second training data sample set; and inputting the first training data sample set and the second training data sample set into the misapplication prediction model to train the misapplication prediction model.

[0012] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication includes extracting data features from data samples of normal vehicle driving and data samples of accelerator pedal misapplication to obtain a first training data sample set, which includes selecting data features of the moments before and after the change in accelerator pedal opening amount is greater than a preset change amount within a predetermined time period as the first training data sample set, wherein the data features are associated with the vehicle driving data.

[0013] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication is wherein the misapplication prediction model is trained based on the XGBoost algorithm.

[0014] According to one embodiment or any of the above embodiments of the present invention, the method for detecting accelerator pedal misapplication is wherein the model parameters of the misapplication prediction model are dynamically adjusted based on driver behavior data.

[0015] According to a second aspect of the present invention, a system for detecting accidental accelerator pedal press is provided, comprising: a data acquisition unit configured to acquire vehicle driving data; a processing unit configured to process the vehicle driving data using an accidental press prediction model to predict whether an accelerator pedal press has occurred; and a judgment unit configured to: determine whether the vehicle is in a predetermined acceleration scenario in response to the prediction that the accelerator pedal has been pressed accidentally; and determine that the accelerator pedal press has occurred accidentally in response to the determination that the vehicle is not in the predetermined acceleration scenario.

[0016] According to an embodiment of the present invention, a control system for detecting accelerator pedal misapplication is provided, wherein the vehicle driving data includes one or more of the following: vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening.

[0017] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication, wherein the processing unit is further configured to: input the vehicle driving data into the misapplication prediction model to generate a misapplication probability value; and predict that the accelerator pedal has been misapplication when the generated misapplication probability value is greater than a preset probability value.

[0018] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication includes one or more of the following predetermined acceleration scenarios: uphill acceleration, acceleration exiting a curve, acceleration during forward / reverse gear shifting, and acceleration after brake release.

[0019] According to one embodiment or any of the above embodiments of the present invention, a control system for detecting accelerator pedal misapplication is provided, wherein the system further includes a selection unit configured to: determine the current road scene of the vehicle based on vehicle GPS data before processing the vehicle driving data using a misapplication prediction model to predict whether the accelerator pedal has been misapplication; and select a misapplication prediction model associated with the determined current road scene of the vehicle.

[0020] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication is trained by the following steps: extracting data features from data samples of normal vehicle driving and sample data of accelerator pedal misapplication to obtain a first training data sample set; performing negative sample augmentation on the sample data of accelerator pedal misapplication to obtain a second training data sample set; and inputting the first training data sample set and the second training data sample set into the misapplication prediction model to train the misapplication prediction model.

[0021] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication includes extracting data features from data samples of normal vehicle driving and data samples of accelerator pedal misapplication to obtain a first training data sample set, which includes selecting data features of the moments before and after the change in the opening amount of the accelerator pedal is greater than a preset change amount within a predetermined time period as the first training data sample set, wherein the data features are associated with the vehicle driving data.

[0022] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication is wherein the misapplication prediction model is trained based on the XGBoost algorithm.

[0023] According to one embodiment or any of the above embodiments of the present invention, the control system for detecting accelerator pedal misapplication is wherein the model parameters of the misapplication prediction model are dynamically adjusted based on driver behavior data.

[0024] According to a third aspect of the present invention, a computer storage medium is provided, the computer storage medium including instructions that, when executed, perform the steps of the method for detecting accidental depressing of the accelerator pedal according to a first aspect of the present invention.

[0025] According to a fourth aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method for detecting accidental accelerator pedal depressing according to a first aspect of the present invention.

[0026] According to a fifth aspect of the invention, a vehicle is provided, the vehicle including a system for detecting accidental pressing of the accelerator pedal according to a second aspect of the invention.

[0027] The scheme for detecting accelerator pedal misapplication according to one or more embodiments of the present invention can establish a misapplication prediction model by combining sample data of actual accelerator pedal misapplication. By improving the selection process of model training samples and the adjustment process of model parameters, the accuracy and reliability of the model prediction results are improved. In addition, when the misapplication prediction model predicts that the accelerator pedal has been misapplication, it further determines whether the vehicle is in a special acceleration scenario, which further improves the robustness of accelerator pedal misapplication detection and avoids misjudging normal driving behavior in special acceleration scenarios as misapplication of the accelerator pedal and intervening accordingly, thereby improving the driving experience. Attached Figure Description

[0028] The above and / or other aspects and advantages of the present invention will become clearer and more readily understood from the following description taken in conjunction with the accompanying drawings, in which the same or similar elements are denoted by the same reference numerals. In the drawings:

[0029] Figure 1 A flowchart of a method for detecting accidental accelerator pedal depressing according to an embodiment of the present invention is shown.

[0030] Figure 2 A schematic diagram of a system for detecting accidental accelerator pedal presses according to an embodiment of the present invention is shown.

[0031] Figure 3 A schematic diagram of a process for establishing a step-by-step prediction model according to an embodiment of the present invention is shown.

[0032] Figure 4 A block diagram of a computer device according to an embodiment of the present invention is shown. Detailed Implementation

[0033] The following detailed description is merely exemplary in nature and is not intended to limit the disclosed technology or its application and use. Furthermore, it is not intended to be bound by any express or implied theory presented in the foregoing technical fields, background art, or the following detailed description.

[0034] In the following detailed description of the embodiments, numerous specific details are set forth in order to provide a more thorough understanding of the disclosed technology. However, it will be apparent to those skilled in the art that the disclosed technology can be practiced without these specific details. In other instances, well-known features have not been described in detail to avoid unnecessarily complicating the description.

[0035] Terms such as "comprising" and "including" indicate that, in addition to the units and steps that are directly and explicitly stated in the specification, the technical solution of the present invention does not exclude the presence of other units and steps that are not directly or explicitly stated. Terms such as "first" and "second" do not indicate the order of the units in terms of time, space, size, etc., but are merely used to distinguish the units.

[0036] In the following, various exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0037] Figure 1 A flowchart of a method for detecting accidental accelerator pedal depressing according to an embodiment of the present invention is shown.

[0038] like Figure 1 As shown, in step 110, vehicle driving data is acquired. Optionally, the vehicle driving data may include, but is not limited to, vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening.

[0039] For example, vehicle speed can be obtained through an electronic stability control system, steering wheel angle can be obtained through a steering wheel angle sensor, longitudinal acceleration can be obtained through an inertial measurement unit, and accelerator pedal opening can be directly obtained by collecting the voltage of the accelerator pedal sensor.

[0040] In step 120, the acquired vehicle driving data is processed using a mis-application prediction model to predict whether the accelerator pedal has been mis-application. This will be discussed in conjunction with... Figure 3 The training process of the accelerator pedal misapplication prediction model is described in detail. In the context of this invention, the accelerator pedal misapplication prediction model is established at least in part based on sample data of actual accelerator pedal misapplication, and the selection process of other model training samples and the adjustment process of model parameters are further improved, thereby improving the accuracy and reliability of the model prediction results.

[0041] Optionally, in step 120, vehicle driving data can be used as input to the accelerator pedal misapplication prediction model to generate a misapplication probability value. When the generated misapplication probability value is greater than a preset probability value, it is predicted that the accelerator pedal has been misapplication. As an example, the preset probability value can be given by the accelerator pedal misapplication prediction model obtained from the initial training and updated synchronously during subsequent iterations of the accelerator pedal misapplication prediction model.

[0042] If a mis-pressing of the accelerator pedal is predicted, proceed to step 130. In step 130, it is determined whether the vehicle is in a predetermined acceleration scenario. For example, a predetermined acceleration scenario may include, but is not limited to, uphill acceleration, acceleration exiting a curve, acceleration during gear shifting (forward / reverse), and acceleration after brake release. It is understood that a predetermined acceleration scenario in this paper refers to a vehicle driving scenario where the vehicle's driving data is highly similar to the vehicle's driving data in the event of a mis-pressing of the accelerator pedal; the predetermined acceleration scenarios listed above are merely illustrative. Therefore, if the prediction result of the mis-pressing prediction model indicates that the accelerator pedal has been mis-pressed, it is necessary to further determine whether the vehicle is in a predetermined acceleration scenario. If it is determined that the vehicle is not in a predetermined acceleration scenario, it is confirmed that the accelerator pedal has been mis-pressed. This improves the robustness and effectiveness of detecting mis-pressing of the accelerator pedal, avoiding misjudging normal driving behavior in special acceleration scenarios as mis-pressing of the accelerator pedal, which could severely impact the driving experience. When it is determined that the accelerator pedal has been mis-pressed, intervention can be performed using a response strategy specified within the vehicle controller. For example, the response strategy may include, but is not limited to, audible warnings, tactile warnings via pedal reaction force, disabling torque output, and emergency braking.

[0043] For example, in step 130, the longitudinal acceleration obtained by the vehicle's inertial measurement unit can be used to determine whether the vehicle is in an uphill acceleration scenario, the turn signal can be used to assist in determining whether the vehicle is in a curve exit acceleration scenario, the gear position signal can be used to determine whether the vehicle is in a forward / reverse gear shift acceleration scenario, and the brake pedal status signal can be used to determine whether the vehicle is in a brake release acceleration scenario.

[0044] In one embodiment, prior to step 120, the current road scenario of the vehicle can be determined based on vehicle GPS data, and a mis-pedal prediction model associated with the determined road scenario can be selected. Exemplarily, the road scenario can include, but is not limited to, urban roads, highways, national roads, provincial roads, county roads, rural roads, parking lot roads, etc. In one embodiment, the current road scenario of the vehicle can be determined based on vehicle GPS data, and a response strategy associated with the determined road scenario can be selected to intervene in the driver's mis-pedaling of the accelerator pedal when it is determined that a mis-pedaling has occurred. For example, when it is determined that the vehicle is in a parking lot road, an audible warning signal can be used to alert the driver that a mis-pedaling has occurred; when it is determined that the vehicle is in an urban road, a tactile feedback of the pedal reaction force can be used to alert the driver that a mis-pedaling has occurred.

[0045] By selecting a mis-application prediction model associated with the vehicle's current road scenario to process vehicle driving data and predict whether the accelerator pedal has been mis-applied, the model can be better applied to different road scenarios, thereby improving the accuracy and reliability of the model's prediction results. Furthermore, by selecting a response strategy associated with the determined current road scenario to intervene in the driver's mis-application of the accelerator pedal when it is determined to have occurred, intervention in the driver's driving behavior can be minimized, improving the driving experience.

[0046] The method for detecting accelerator pedal misapplication according to one aspect of the present invention can further determine whether the vehicle is in a special acceleration scenario when the misapplication prediction model predicts that the accelerator pedal has been misapplication, thereby improving the robustness of accelerator pedal misapplication detection and avoiding intervention due to misjudging normal driving behavior in special acceleration scenarios as misapplication of the accelerator pedal, thus improving the driving experience.

[0047] Figure 2 A schematic diagram of a system for detecting accidental accelerator pedal presses according to an embodiment of the present invention is shown.

[0048] like Figure 2 As shown, the system 200 for detecting accidental pressing of the accelerator pedal includes a data acquisition unit 210, a processing unit 220, and a judgment unit 230.

[0049] The acquisition unit 210 can be configured to acquire vehicle driving data. Optionally, the vehicle driving data may include, but is not limited to, vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening.

[0050] For example, the acquisition unit 210 can acquire vehicle speed through the electronic stability controller system, steering wheel angle through the steering wheel angle sensor, longitudinal acceleration through the inertial measurement unit, and accelerator pedal opening directly by acquiring the voltage of the accelerator pedal sensor.

[0051] Processing unit 220 can be configured to process the acquired vehicle driving data using a mis-application prediction model to predict whether the accelerator pedal has been mis-application. This will be discussed in conjunction with... Figure 3 Describe in detail the training process of the accidental step prediction model.

[0052] Optionally, the processing unit 220 can be configured to use vehicle driving data as input to the accelerator pedal misapplication prediction model to generate a misapplication probability value. When the generated misapplication probability value is greater than a preset probability value, it predicts that the accelerator pedal has been misapplication. As an example, the preset probability value can be given by the initially trained accelerator pedal misapplication prediction model and updated synchronously during subsequent iterations of the accelerator pedal misapplication prediction model.

[0053] The judgment unit 230 can be configured to determine whether the vehicle is in a predetermined acceleration scenario in response to the processing unit 220 predicting that the accelerator pedal has been accidentally pressed, and to determine that the accelerator pedal has been accidentally pressed in response to the judgment that the vehicle is not in a predetermined acceleration scenario.

[0054] As an example, the predetermined acceleration scenarios may include, but are not limited to, uphill acceleration, acceleration exiting a curve, acceleration during gear shifting (forward or reverse), and acceleration after brake release. It is understood that the predetermined acceleration scenarios in this paper refer to vehicle driving scenarios where the vehicle's driving data is highly similar to the vehicle driving data in the event of a mis-pressed accelerator pedal. The predetermined acceleration scenarios listed above are merely illustrative. Therefore, if the mis-press prediction model predicts that the accelerator pedal has been mis-pressed, it is necessary to further determine whether the vehicle is within the predetermined acceleration scenario. If it is determined that the vehicle is not within the predetermined acceleration scenario, then a mis-press of the accelerator pedal is confirmed. This improves the robustness and effectiveness of detecting mis-pressed accelerator pedals, avoiding misjudging normal driving behavior in special acceleration scenarios as mis-pressed accelerator pedal behavior, which could severely impact the driving experience.

[0055] For example, the judgment unit 230 can be configured to determine whether the vehicle is in an uphill acceleration scenario by using the longitudinal acceleration obtained by the vehicle's inertial measurement unit, to assist in determining whether the vehicle is in a corner exit acceleration scenario by combining the turn signal, to determine whether the vehicle is in a scenario of acceleration when switching between forward and reverse gears by combining the gear signal, and to determine whether the vehicle is in a scenario of acceleration after the brake is released by using the brake pedal status signal.

[0056] In one embodiment, the system 200 for detecting accidental accelerator pedal press may further include a selection unit ( Figure 2 (Not shown in the image) This unit can be configured to determine the current road scenario of the vehicle based on vehicle GPS data before processing vehicle driving data using a mis-application prediction model to predict whether the accelerator pedal has been mis-application, and to select a mis-application prediction model associated with the determined current road scenario. Exemplarily, the road scenario may include, but is not limited to, urban roads, highways, national roads, provincial roads, county roads, rural roads, parking lot roads, etc. In one embodiment, the selection unit can also be configured to determine the current road scenario of the vehicle based on vehicle GPS data and select a response strategy associated with the determined current road scenario to intervene in the driver's mis-application of the accelerator pedal when it is determined that the accelerator pedal has been mis-application. For example, when it is determined that the vehicle is in a parking lot road, an audible warning signal can be used to alert the driver that the accelerator pedal has been mis-application; when it is determined that the vehicle is in an urban road, a tactile feedback of the pedal reaction force can be used to alert the driver that the accelerator pedal has been mis-application.

[0057] By selecting a mis-application prediction model associated with the vehicle's current road scenario to process vehicle driving data and predict whether the accelerator pedal has been mis-applied, the model can be better applied to different road scenarios, thereby improving the accuracy and reliability of the model's prediction results. Furthermore, by selecting a response strategy associated with the determined current road scenario to intervene in the driver's mis-application of the accelerator pedal when it is determined to have occurred, intervention in the driver's driving behavior can be minimized, improving the driving experience.

[0058] The system for detecting accelerator pedal misapplication according to one aspect of the present invention can further determine whether the vehicle is in a special acceleration scenario when the misapplication prediction model predicts that the accelerator pedal has been misapplication, thereby improving the robustness of accelerator pedal misapplication detection and avoiding intervention due to misjudging normal driving behavior in special acceleration scenarios as misapplication of the accelerator pedal, thus improving the driving experience.

[0059] Figure 3 A schematic diagram illustrating the process for establishing a model for predicting accidental footsteps according to an embodiment of the present invention is shown. Figure 3As shown, the process of building a model for predicting accidental stepping can include three stages: data sample collection and feature extraction, model training, and model output.

[0060] Box 310 illustrates the data sample acquisition and feature extraction stage for establishing the accelerator pedal misapplication prediction model. Optionally, in this stage, data features can be extracted from data samples of normal vehicle driving and samples of accelerator pedal misapplication to obtain a first training data sample set. Negative sample augmentation can be performed on the accelerator pedal misapplication samples to obtain a second training data sample set. The first and second training data sample sets are then used as training data for the accelerator pedal misapplication prediction model. By including accelerator pedal misapplication samples as part of the training data, the trained model can better adapt to the data characteristics of actual accelerator pedal misapplication situations, ensuring the effectiveness of the model's prediction results.

[0061] Optionally, extracting data features from data samples of normal vehicle operation and accelerator pedal misapplication to obtain a first training data sample set may include selecting data features from moments before and after a predetermined time period in which the change in accelerator pedal opening exceeds a preset change amount as the first training data sample set. For example, the predetermined time period may be set to 1 second, the preset change amount may be set to 70%, and data features from moments before and after a change in accelerator pedal opening exceeding 70% within 1 second may be selected as the first training data sample set. Exemplarily, the predetermined time period can be adjusted through calibration. Exemplarily, the extracted data features may include, but are not limited to, vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and accelerator pedal opening change rate.

[0062] In one embodiment, data sample collection and feature extraction may include collecting multi-vehicle driving sample data, collecting time sequence data before and after a single vehicle's accelerator pedal mis-pressing, data feature extraction, and negative sample data augmentation.

[0063] For example, vehicle operation data that meets the filtering criteria can be collected from a database of operating vehicles as multi-vehicle driving sample data, where the filtering criteria can be determined based on big data statistical results. For instance, in most cases before and after an accidental accelerator pedal press, the accelerator pedal opening is greater than 90%, and the rate of change of the accelerator pedal opening within 1 second is greater than 20%. Therefore, this can be used as a filtering criterion to select normal driving data of vehicles in the database that meet this criterion as multi-vehicle driving sample data.

[0064] For example, time-series data of the corresponding time segment can be collected from the vehicle database based on the time of the accelerator pedal event as the time sequence data before and after the accidental pressing of the accelerator pedal in a single vehicle.

[0065] For example, data feature extraction includes data segmentation and feature extraction operations on multi-vehicle driving sample data and accelerator pedal mis-pressing sample data. For instance, data features can be extracted for the moment when the accelerator pedal opening changes from less than 20% of the initial value to more than 90% of the opening within 1 second.

[0066] For example, negative sample data augmentation involves performing negative sample augmentation on the time-series data before and after the accelerator pedal mis-pressing to increase the number of mis-pressing data samples. For instance, negative sample augmentation can be performed by changing the steering wheel angle of the mis-pressing data (where the steering wheel angle is to the left) to the right, and by changing the steering wheel angle of the mis-pressing data (where the steering wheel angle is to the right) to the left. Similarly, other data features can be processed in the same way.

[0067] By performing negative sample augmentation on the accelerator pedal mis-pressing data samples, the problem of limited training samples for mis-pressing data, which is an extremely low-probability event, can be improved, thus affecting the model training results.

[0068] Box 320 illustrates the model training phase. Optionally, in this phase, a model suitable for the data characteristics can be selected for training, while prediction accuracy is used as the direction for model optimization. For example, the XGBoost algorithm can be selected for model training based on factors such as data size and data distribution characteristics. By using the XGBoost algorithm for model training, the accuracy, efficiency, and flexibility of model training can be improved.

[0069] It should be noted that using the XGBoost algorithm to train a model is merely an example. Other model training algorithms may be selected to train the model without departing from the spirit and scope of this invention.

[0070] Optionally, the model parameters of the accidental accelerator pedal prediction model can be provided by the initially trained accidental accelerator pedal prediction model and updated synchronously during subsequent iterations. For example, the model parameters can be dynamically adjusted based on driver behavior data. For instance, in-vehicle sensors can be embedded to monitor driver behavior and a driver behavior model can be built based on this behavior. Different model parameters can then be set for different drivers based on the results of this driver behavior model to calibrate the model. This calibrated model can accurately determine whether a driver's operation is an accidental accelerator pedal press based on their driving habits. In one embodiment, a model parameter interface for dynamically adjusting model parameters can be reserved for the accidental accelerator pedal prediction model, allowing for dynamic adjustment of model parameters according to different drivers' driving habits.

[0071] Box 330 shows the model output. Optionally, the model output can indicate the probability of accelerator pedal misapplication.

[0072] The scheme for training a mis-pedal prediction model according to one or more embodiments of the present invention can combine actual accelerator pedal mis-pedal sample data, and improve the accuracy and reliability of the model prediction results by improving the selection process of model training samples and the adjustment process of model parameters.

[0073] Figure 4 A block diagram of a computer device according to an embodiment of the present invention is shown. Figure 4 As shown, the computer device 400 includes a memory 410, a processor 420, and a computer program 430 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 430, it implements, for example... Figure 1 The steps of a method for detecting accidental accelerator pedal depressing according to one aspect of the invention are shown.

[0074] Additionally, as described above, the present invention can also be implemented as a computer storage medium storing a program for causing a computer to execute a method for detecting accidental accelerator pedal depressing according to one aspect of the present invention.

[0075] Here, computer storage media can be various types, such as disks (e.g., hard disks, optical disks, etc.), cards (e.g., memory cards, optical cards, etc.), semiconductor memory (e.g., ROM, non-volatile memory, etc.), and tapes (e.g., magnetic tape, cassette tape, etc.).

[0076] Where applicable, the various embodiments provided by this disclosure may be implemented using hardware, software, or a combination of hardware and software. Furthermore, where applicable, without departing from the scope of this disclosure, the various hardware and / or software components described herein may be combined into composite components comprising software, hardware, and / or both. Where applicable, without departing from the scope of this disclosure, the various hardware and / or software components described herein may be divided into sub-components comprising software, hardware, or both. Additionally, where applicable, it is contemplated that software components may be implemented as hardware components, and vice versa.

[0077] The software (such as program code and / or data) according to this disclosure can be stored on one or more computer storage media. It is also contemplated that the software identified herein can be implemented using one or more networked and / or otherwise general-purpose or special-purpose computers and / or computer systems. Where applicable, the order of the various steps described herein can be changed, combined into compound steps, and / or divided into sub-steps to provide the features described herein.

[0078] The embodiments and examples presented herein are provided to best illustrate embodiments of the invention and its particular applications, thereby enabling those skilled in the art to practice and use the invention. However, those skilled in the art will understand that the above description and examples are provided merely for ease of illustration and example. The descriptions presented are not intended to cover all aspects of the invention or to limit the invention to the precise forms disclosed.

Claims

1. A method for detecting accidental pressing of the accelerator pedal, characterized in that, The method includes: Acquire vehicle driving data, wherein the vehicle driving data includes one or more of the following: vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening; The vehicle driving data is processed using a mis-pressing prediction model to predict whether the accelerator pedal is mispressed. The process includes inputting the vehicle driving data into the mis-pressing prediction model to generate a mis-pressing probability value, and predicting that the accelerator pedal is mispressed when the generated mis-pressing probability value is greater than a preset probability value. In response to predicting a misapplication of the accelerator pedal, a determination is made as to whether the vehicle is in a predetermined acceleration scenario, wherein the predetermined acceleration scenario includes one or more of the following: uphill acceleration, acceleration exiting a corner, acceleration during gear shifting (forward or reverse), and acceleration after brake release; and In response to determining that the vehicle is not in the predetermined acceleration scenario, it is determined that the accelerator pedal has been mistakenly pressed; The accidental step prediction model is trained through the following steps: Data features are extracted from data samples of normal vehicle operation and data samples of accidental accelerator pedal press to obtain the first training data sample set; Negative sample augmentation is performed on the sample data of the accelerator pedal mis-pressing to obtain a second training data sample set; and The first training data sample set and the second training data sample set are input into the accidental step prediction model to train the accidental step prediction model, wherein the accidental step prediction model is trained based on the XGBoost algorithm, and the model parameters of the accidental step prediction model are dynamically adjusted based on driver behavior data.

2. The method according to claim 1, wherein the method further comprises: Before using the accidental accelerator pedal prediction model to process the vehicle driving data to predict whether the accelerator pedal has been accidentally pressed, the current road scene of the vehicle is determined based on the vehicle's GPS data. as well as Select a misstep prediction model associated with the determined road scenario in which the vehicle is currently located.

3. The method according to claim 1, wherein extracting data features from data samples of normal vehicle driving and accelerator pedal mis-pressing to obtain a first training data sample set includes: The data features at the moments before and after the change in accelerator pedal opening is greater than a preset change within a predetermined time period are selected as the first training data sample set, wherein the data features are associated with the vehicle driving data.

4. A system for detecting accidental pressing of the accelerator pedal, characterized in that, The system includes: The acquisition unit is configured to acquire vehicle driving data, wherein the vehicle driving data includes one or more of the following: vehicle speed, steering wheel angle, longitudinal acceleration, accelerator pedal opening, and rate of change of accelerator pedal opening. A processing unit configured to process the vehicle driving data using a mis-application prediction model to predict whether the accelerator pedal has been mis-application, wherein the processing unit is further configured to input the vehicle driving data into the mis-application prediction model to generate a mis-application probability value, and to predict that the accelerator pedal has been mis-application when the generated mis-application probability value is greater than a preset probability value; and The decision unit is configured as follows: In response to predicting a misapplication of the accelerator pedal, a determination is made as to whether the vehicle is in a predetermined acceleration scenario, wherein the predetermined acceleration scenario includes one or more of the following: uphill acceleration, acceleration exiting a corner, acceleration during gear shifting (forward or reverse), and acceleration after brake release; and In response to determining that the vehicle is not in the predetermined acceleration scenario, it is determined that the accelerator pedal has been accidentally pressed; The accidental step prediction model is trained through the following steps: Data features are extracted from data samples of normal vehicle operation and data samples of accidental accelerator pedal press to obtain the first training data sample set; Negative sample augmentation is performed on the sample data of the accelerator pedal mis-pressing to obtain a second training data sample set; and The first training data sample set and the second training data sample set are input into the accidental step prediction model to train the accidental step prediction model, wherein the accidental step prediction model is trained based on the XGBoost algorithm, and the model parameters of the accidental step prediction model are dynamically adjusted based on driver behavior data.

5. The system of claim 4, wherein the system further comprises a selection unit configured to: Before processing the vehicle driving data using the accelerator pedal misapplication prediction model to predict whether the accelerator pedal has been misapplication, the current road scenario of the vehicle is determined based on the vehicle's GPS data; and Select a misstep prediction model associated with the determined road scenario in which the vehicle is currently located.

6. The system according to claim 4, wherein extracting data features from data samples of normal vehicle driving and accelerator pedal mis-pressing to obtain a first training data sample set includes: The data features at the moments before and after the change in accelerator pedal opening is greater than a preset change within a predetermined time period are selected as the first training data sample set, wherein the data features are associated with the vehicle driving data.

7. A computer storage medium, characterized in that, The computer storage medium includes instructions that, when executed, perform the method according to any one of claims 1 to 3.

8. A computer device, characterized in that, The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the method of any one of claims 1 to 3.

9. A vehicle, characterized in that, The system comprising any one of claims 4 to 6 for detecting accidental pressing of the accelerator pedal.

Images