False stepping judgment method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of Internet of Vehicles, and provides a mistaken stepping judgment method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining current driving data which comprises at least one of vehicle position road information, vehicle driving information and environment weather information, and driving related information; based on the current driving data and a mistaken stepping judgment model corresponding to the driver, the accelerator stepping operation of the driver is judged, and the mistaken stepping judgment model is obtained based on historical driving data training of the driver. According to the mistaken stepping judgment method and device, the electronic equipment, the storage medium and the mistaken stepping judgment model, the personal driving habits of the driver can be reflected, individuation and pertinence are better achieved, correct judgment can be effectively made according to the different driving habits, whether the accelerator is mistakenly stepped or not can be predicted more accurately, and the driving experience of the driver is improved. And the accuracy of mistaken stepping judgment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle networking, and in particular to a method, device, electronic device and storage medium for judging mis-stepping. Background Art

[0002] Cars have become an indispensable means of transportation in people's daily lives. Their intelligence is constantly improving, and the safety of car driving is also related to the safety of people's lives and property. In recent years, there have been frequent reports of drivers mistakenly stepping on the accelerator instead of the brake in special scenarios, causing major safety accidents.

[0003] At present, most of the solutions for accidental accelerator pedaling are passive suppression by the system, which collects the driver's driving parameters and compares them with the configured parameter thresholds to determine whether the accelerator pedal is accidentally pressed, thereby suppressing acceleration. However, different drivers have different driving habits, and the universal configuration parameter thresholds are relatively fixed, resulting in poor accuracy in the judgment of accidental pedaling. Summary of the invention

[0004] The present invention provides a method, device, electronic device and storage medium for judging whether the driver has stepped on the accelerator by mistake, so as to solve the defect of poor accuracy in judging whether the driver has stepped on the accelerator by mistake in the prior art.

[0005] The present invention provides a method for judging an accidental stepping, comprising: Acquiring current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information; The accelerator pedal operation of the driver is judged based on the current driving data and a wrong pedal judgment model corresponding to the driver, wherein the wrong pedal judgment model is obtained by training based on the driver's historical driving data.

[0006] According to the method for judging the accidental stepping provided by the present invention, the step of obtaining the accidental stepping judgment model comprises: Extracting features of each information dimension from the driver's historical driving data to obtain features of each dimension, and determining a baseline feature of the driver based on the features of each dimension; Based on the differences between the features in each dimension and the reference features, the accelerator operation of the driver is predicted, and the mis-accelerator judgment model is trained based on the prediction results.

[0007] According to the method for judging the mis-operation of the pedal provided by the present invention, the prediction of the driver's accelerator operation based on the difference between the features of each dimension and the reference features includes: Predicting the driver's accelerator operation based on the information gain weight of each dimensional feature and the difference between each dimensional feature and the reference feature; The information gain weight of any dimensional feature is determined based on the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is included, and the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is not included.

[0008] According to the method for judging the mis-operation of the pedal provided by the present invention, the prediction of the driver's accelerator operation based on the difference between the features of each dimension and the reference features includes: Fusing the features of each dimension to obtain a fused feature; Based on the difference between the features in each dimension and the reference feature, and the difference between the fusion feature and the reference feature, the accelerator operation of the driver is predicted.

[0009] According to the method for judging the mis-stepping of the vehicle provided by the present invention, the determining of the baseline characteristics of the driver based on the characteristics of each dimension includes: The reference feature is determined based on an average value of features in each dimension corresponding to the case where the driver steps on the accelerator correctly.

[0010] According to the method for judging the mis-stepping of the accelerator pedal provided by the present invention, the accelerator pedal operation of the driver is judged, and then the method further comprises: If the judgment result indicates that the accelerator pedal operation is an erroneous operation, a reminder message is sent.

[0011] According to the method for judging the mistaken stepping provided by the present invention, the method further includes: Identifying the driver; If the identity recognition result indicates that the driver is a new user, constructing a wrong-stepping judgment model corresponding to the driver based on the current driving data; If the identity recognition result indicates that the driver is not a new user, a mistaken stepping judgment model corresponding to the driver is loaded, and the mistaken stepping judgment model is updated based on current driving data.

[0012] The present invention also provides a mis-stepping judgment device, comprising: A data acquisition unit, used to acquire current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information; The mis-operation judgment unit is used to judge the accelerator operation of the driver based on the current driving data and a mis-operation judgment model corresponding to the driver, wherein the mis-operation judgment model is trained based on the driver's historical driving data.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned methods for determining an incorrect stepping on a device is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, any of the above-mentioned methods for judging an incorrect stepping on a vehicle is implemented.

[0015] The present invention also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for judging mis-stepping.

[0016] The method, device, electronic device and storage medium for judging the accidental stepping of the accelerator provided by the present invention obtain an accidental stepping judgment model corresponding to the driver by collecting the driver's historical driving data for training. The accidental stepping judgment model can reflect the driver's personal driving habits. Compared with the general configuration parameter thresholds used in related technologies, the accidental stepping judgment model in this embodiment is more personalized and targeted, and can effectively make correct judgments based on different driving habits. Therefore, it can more accurately predict whether the accelerator is accidentally stepped on, thereby improving the accuracy of the accidental stepping judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 This is one of the flow charts of the method for judging mis-stepping provided by the present invention.

[0019] Figure 2 This is one of the flow charts of the method for acquiring the mis-stepping judgment model provided by the present invention.

[0020] Figure 3 This is the second flow chart of the method for acquiring the mis-stepping judgment model provided by the present invention.

[0021] Figure 4 It is a structural schematic diagram of the prediction module provided by the present invention.

[0022] Figure 5 This is the second flow chart of the method for judging the mis-stepping provided by the present invention.

[0023] Figure 6 It is a structural schematic diagram of the mis-stepping judgment device provided by the present invention.

[0024] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] In recent years, there have been frequent reports of drivers mistakenly stepping on the accelerator instead of the brake in special scenarios, causing major safety accidents. Therefore, it is of great significance to use monitoring equipment and big data models to determine whether the driver has mistakenly stepped on the accelerator, and to promptly remind the driver when an erroneous driving operation occurs, thereby reducing the probability of accidents.

[0027] At present, most of the solutions for handling accidental accelerator pedal pressing are system passive suppression, which collects the driver's driving parameters and compares them with the configured parameter thresholds to determine whether the accelerator pedal is pressed accidentally, thereby suppressing acceleration. The existing technical solutions do not have the driver's personalized data. Because different drivers have different driving habits, and the universal configuration parameter thresholds are fixed, it is impossible to effectively make correct judgments based on different driving habits, resulting in poor accuracy in the judgment of accidental pedal pressing.

[0028] In response to the above problems, an embodiment of the present invention proposes a method for judging the accelerator pedal misoperation. In the method, current driving data is obtained during driving, and the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental meteorological information, and driving-related information; based on the current driving data and the accelerator pedal misoperation judgment model corresponding to the driver, the accelerator pedal misoperation operation of the driver is judged. The accelerator pedal misoperation judgment model is trained based on the driver's historical driving data, and the accelerator pedal misoperation judgment model can reflect the driver's personal driving habits. Compared with the general configuration parameter threshold used in the related technology, the accelerator pedal misoperation judgment model in this embodiment is more personalized and targeted, and can effectively make correct judgments based on different driving habits, thereby more accurately predicting whether the accelerator pedal is misoperated, and improving the accuracy of the accelerator pedal misoperation judgment.

[0029] The embodiment of the present invention can be applied to the scene where it is necessary to judge whether the driver mistakenly stepped on the accelerator instead of the brake during the driving process of the car. The execution subject of the method can be an electronic device such as a terminal device, a computer, a server, a server cluster or a specially designed mistaken stepping judgment device, or it can be a mistaken stepping judgment device set in the electronic device, and the mistaken stepping judgment device can be implemented by software, hardware or a combination of the two.

[0030] Figure 1 FIG. 1 is one of the flow charts of the method for judging the misstepping provided by the present invention. Figure 1 As shown, the method comprises the following steps: Step 110, obtaining current driving data, the current driving data including at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information.

[0031] Specifically, the current driving data can be collected when the driver is detected to have stepped on the accelerator or brake hard. The vehicle position road information refers to the current position of the vehicle and the road information at that position, describing the factors affecting whether the driver has stepped on the accelerator by mistake in the spatial dimension. The vehicle position is obtained in real time using a positioning device built into or installed after the vehicle, such as a GPS locator. The vehicle position can be expressed in latitude and longitude, for example.

[0032] The road information at the vehicle's location represents the vehicle's driving environment and covers various data related to the road. The road information may include the type of the current road, road conditions, traffic rules, etc. The type of the current road may include, for example, expressways, trunk roads, provincial roads, county roads, village roads, mountain roads, etc. The road condition refers to the traffic situation of the current road, for example, the navigation map information in the vehicle can be used to obtain whether the current road section is congested, and whether there are obstacles within the driving range through car image recognition; traffic rules refer to the traffic regulations of the current road, such as whether there is a speed limit, the speed limit threshold, etc., which can be obtained through the navigation map information.

[0033] Vehicle driving information is used to describe the dynamic performance or state of the vehicle during driving, and to describe the factors that affect the driver's accidental accelerator pedaling from the vehicle dimension. Vehicle driving information may include current vehicle speed, average vehicle speed, speed range for the current time period, braking system status, vehicle acceleration, etc. Vehicle driving information can be read through the vehicle's OBD-II (On-Board Diagnostics) port.

[0034] Environmental meteorological information is used to describe the factors that affect whether the driver accidentally steps on the accelerator within the scope of the external environment. Environmental meteorological information refers to the state of the natural environment or atmospheric environment at the current moment, which may include the current weather, temperature, season, time, visibility, etc.

[0035] Driving and riding related information is used to describe the factors that affect whether the driver accidentally steps on the accelerator in the dimension of personalization. Driving and riding related information includes basic information of the driver and feedback information from fellow passengers. Basic information of the driver may include, for example, the driver's identification code, gender, driving experience, age, etc. The basic information of the driver can be stored in advance and the current driver's information can be obtained through identity recognition and matching. Feedback information from fellow passengers may include whether the fellow passengers have obvious voice keywords (such as "brake", "stop stop", "scream") prompts, which can be collected through the sound pickup equipment in the car.

[0036] In this embodiment, the data used to determine the possibility of the driver's mistaken stepping on the brake pedal includes at least one of the vehicle's position and road information, vehicle driving information, and environmental weather information, as well as driving-related information.

[0037] Step 120 , based on the current driving data and a mis-operation judgment model corresponding to the driver, the driver's accelerator operation is judged. The mis-operation judgment model is trained based on the driver's historical driving data.

[0038] Specifically, the current driving data is input into the mistaken step judgment model, which processes the current driving data and predicts whether the driver's current driving operation is a mistaken step on the accelerator based on the data processing results. Misstepping on the accelerator means that the model predicts that the brake should be stepped on in the current scenario, but the actual driving operation of the driver is to step on the accelerator.

[0039] Before executing step 120, a large amount of historical driving data of the driver can be collected and trained by a machine learning algorithm. The historical driving data can be used to preliminarily construct driving habit models of different drivers. For example, user 001 is 35 years old, has 9 years of driving experience, and has mature driving skills. In good weather, high visibility, and pleasant temperature at noon, on a provincial road without obstacles with a speed limit of 60km, the user's habitual driving speed range is 45~55km / h, and the user's habitual acceleration range is ±3km / s. If the deviation distance between the collected feature data and the user's mis-stepping judgment model data exceeds the threshold, it is judged that the user may have mistakenly stepped on the accelerator. The more feature data collected, the wider the user scenario covered, and after model training, the more accurately the user's accelerator operation can be predicted.

[0040] During the training process, the accidental step judgment model can learn the driver's personal driving habits from historical driving data, that is, in what scenarios should the brake be pressed and in what scenarios should the accelerator be pressed. The accidental step judgment model thus obtained can reflect the driver's personal driving habits and is more personalized and targeted.

[0041] The mis-accelerator judgment model can reflect the driver's benchmark driving data when the driver presses the accelerator correctly. If the difference between the current driving data and the benchmark driving data is greater, the probability that the driver's accelerator operation is a mis-accelerator operation is greater; conversely, if the difference between the current driving data and the benchmark driving data is smaller, the probability that the driver's accelerator operation is a mis-accelerator operation is smaller.

[0042] The method provided in an embodiment of the present invention obtains a mis-stepping judgment model corresponding to the driver by collecting the driver's historical driving data for training. The mis-stepping judgment model can reflect the driver's personal driving habits. Compared with the general configuration parameter thresholds used in related technologies, the mis-stepping judgment model in this embodiment is more personalized and targeted, and can effectively make correct judgments based on different driving habits. This can more accurately predict whether the accelerator is mistakenly stepped on, thereby improving the accuracy of mis-stepping judgment.

[0043] Based on any of the above embodiments, Figure 2 is one of the flow charts of the method for obtaining the misstepping judgment model provided by the present invention. Figure 2 As shown, the steps for obtaining the mis-stepping judgment model include: Step 210, extracting features of each information dimension from the driver's historical driving data to obtain features of each dimension, and determining the driver's baseline features based on the features of each dimension; Step 220 , based on the difference between the features of each dimension and the reference features, predict whether the driver's accelerator operation is correct, and train a mis-accelerator judgment model based on the prediction result.

[0044] Specifically, after collecting the driver's historical driving data, the historical driving data can be cleaned, such as removing duplicates, filling missing values, correcting obviously erroneous data, and performing preliminary verification of the data's completeness and accuracy to ensure that the data conforms to the expected format and range.

[0045] Then, the driver's historical driving data is subjected to feature extraction of each information dimension. Here, each information dimension may include four dimensions: vehicle location road information, vehicle driving information, environmental weather information, and driving-related information. The features of each dimension obtained by feature extraction may include vehicle location road features, vehicle driving features, environmental weather features, and driving-related features.

[0046] In some embodiments, the features of each dimension can be represented by digital codes. For the features of driving-related dimensions, for example, through driver authorization, the driver's facial data is obtained, and different numbers are assigned to different users to distinguish users, such as 001; the user's gender is identified by 0 (female) and 1 (male); the user's driving experience and age are identified by numbers, such as 9, 35; collect whether there are obvious voice keywords (such as "brake", "stop stop", "scream") prompts from fellow passengers, with 0 indicating no and 1 indicating yes. In summary, the feature vector of driving-related features can be expressed as [001 (user number), 1 (male), 9 (driving experience), 35 (age), 0 (no voice prompts from fellow passengers), 0 (correct accelerator pedal)].

[0047] For the characteristics of the vehicle position road dimension, such as the driving position of the car, the latitude and longitude information is expressed as (114.33, 30.35); judge the nature of the current road, with 0 representing the expressway, 1 representing the provincial road, 2 representing the provincial road, 3 representing the county road, 4 representing the rural road, and 5 representing other roads; identify whether there are obstacles within the driving range through the car image, with 1 indicating the presence of obvious obstacles and 0 indicating the absence of obstacles; obtain the traffic regulations of the current section through the navigation map information in the car computer, whether there is a speed limit, such as 60km / h. In summary, the vehicle position road feature vector is expressed as [001 (user number), 114.3321 (latitude), 30.35216 (longitude), 1 (provincial road), 0 (no obstacles), 1 (speed limit), 60km / h (speed limit threshold), 0 (correct accelerator)].

[0048] For the characteristics of vehicle driving dimensions, the current vehicle speed is read through the vehicle OBD-II (On-Board Diagnostics) port, such as 55km / h; the average vehicle speed is such as 60km / h; the speed range of the current time period, such as 50km / h in the left range and 70km / h in the right range (indicating that during this period, the driver's driving speed range is 50~70km / h); the state information of the braking system, with 0 indicating the brake and 1 indicating the throttle, such as 1; the acceleration information of the vehicle is obtained through the measurement information of the vehicle acceleration sensor, with positive numbers indicating throttle acceleration and negative numbers indicating brake deceleration, such as 3km / s. In summary, the vehicle driving characteristics are represented by a vector as [001 (user number), 55 (current vehicle speed km / h), 60 (average vehicle speed km / h), 50 (vehicle speed range left range value km / h), 70 (vehicle speed range right range value km / h), 1 (braking state), 3 (indicator acceleration km / s), 0 (correct throttle pedal)].

[0049] For the features of the environmental meteorological dimension, for example, weather conditions can be represented by numbers, with negative numbers indicating bad weather and positive numbers indicating good weather, such as 2; seasonal information can be recorded by month, such as 10; temperature and time can be identified by specific numbers, such as 21, 15; visibility is represented by data, 0 means completely invisible, 100 means completely visible, such as 98. Therefore, the feature vector of the environmental meteorological feature can be expressed as [001 (user number), 2 (good weather), 10 (month), 21 (temperature), 15 (time), 98 (visibility), 0 (correct accelerator pedal)].

[0050] After obtaining the features of each dimension, the driver's baseline features can be determined based on the features of each dimension. The baseline features here refer to the features of the driver when he / she steps on the accelerator correctly. It represents the typical performance of the driver under normal operation, which is equivalent to the driving habit features of each independent user.

[0051] In some embodiments, based on the features of each dimension, determining the baseline features of the driver specifically includes: The benchmark feature is determined based on the average value of the features of each dimension corresponding to the case where the driver steps on the accelerator correctly.

[0052] Each dimension feature in each data record is converted into a corresponding feature vector. These features cover various factors that affect driving behavior. Then the average value of the feature vectors under all correct accelerator pedal conditions is calculated and recorded as the vector corresponding to the baseline feature, which represents the typical performance of the driver under normal operation. The baseline feature corresponds to the driver ID one by one, and each driver has his own independent and personalized baseline feature.

[0053] On this basis, step 220 is performed. By calculating the distance between each dimension feature and the baseline feature vector, the probability of the driver accidentally pressing the accelerator in a specific situation is evaluated. If the distance between the feature vector of a data point and the baseline feature vector significantly exceeds the set threshold, it indicates that the driving behavior may deviate from the normal operating mode and there is a risk of accidentally pressing the accelerator.

[0054] The difference between each dimension feature and the benchmark feature can be measured using chi-square distance, cosine distance, Euclidean distance, etc. In some embodiments, the chi-square measure uses the contingency table analysis method to quantify the degree of difference between two vectors. The principle is to calculate the chi-square statistic. When the chi-square statistic is large, it implies that there is a significant difference in the variable values ​​of the two vectors. The basic calculation formula is formula (1).

[0055] (1) in, yes , Dimensions; is a vector No. The value of latitude, is a vector In the Expected value (average) over dimension; is a vector No. The value of latitude, is a vector In the Expected value on the dimension.

[0056] Since one of the vectors being compared is is the average eigenvector, that is = = , so the chi-square distance formula can be simplified to formula (2).

[0057] (2) In some implementations, based on the difference between the features in each dimension and the reference features, it is predicted whether the driver's accelerator operation is correct, that is, step 220 specifically includes: Based on the information gain weight of each dimension feature and the difference between each dimension feature and the benchmark feature, predict whether the driver's accelerator operation is correct; The information gain weight of any dimensional feature is determined based on the probabilities of the driver pressing the accelerator correctly and incorrectly when any dimensional feature is included, and the probabilities of the driver pressing the accelerator correctly and incorrectly when any dimensional feature is not included.

[0058] Specifically, the chi-square distance is used to characterize the distance between the feature vector of each dimension and the benchmark feature. The smaller the distance, the closer it is to the benchmark feature, that is, the greater the possibility that stepping on the accelerator this time is the correct operation. When using the chi-square distance formula to calculate the distance, the weights of each dimension are the same. In fact, the determining factors for each driver's stepping on the accelerator are different, and the impact of different features on the prediction results should also be different. This embodiment uses information gain to improve the chi-square distance. Before accumulation, each dimension is multiplied by the corresponding weight, which is the information gain of the feature. Information gain is for each feature. For a certain feature, the difference in the overall information entropy when it is included and when it is not included is the amount of information that this feature brings to the whole. For a feature , its information gain The calculation formula is formula (3).

[0059] (3) in, For Category (In this embodiment, the categories are correct accelerator pedal pressing and incorrect accelerator pedal pressing) entropy; It is characterized by When, category The conditional entropy of . Assuming the category have Different values , , ,…, ,but The calculation formula is formula (4).

[0060] (4) in, for The probability of occurrence. According to the conditional probability formula, we can get The calculation formula is formula (5).

[0061] (5) in, for The probability of occurrence, for The probability of not appearing, Features In case of occurrence Probability of occurrence; Features In case of non-appearance Probability of occurrence.

[0062] In summary, features The information gain calculation formula is as follows:

[0063] (6) For the data in this embodiment, the sample category is only and wrong step on the accelerator Two types, , are the probabilities of the driver pressing the accelerator correctly and the driver pressing the accelerator incorrectly when considering this feature. Similarly, , are the probabilities of the driver stepping on the accelerator correctly and incorrectly when this feature is not considered.

[0064] The improved formula combined with the information gain weight is expressed as formula (7).

[0065] (7) in, for The information gain of the corresponding feature subsequence.

[0066] The method provided by the embodiment of the present invention can further improve the accuracy of the prediction by combining the information gain weights of the features of each dimension and the differences between the features of each dimension and the benchmark features to predict whether the driver's accelerator operation is correct.

[0067] In some embodiments, Figure 3 FIG. 2 is a flow chart of the method for obtaining the misstepping judgment model provided by the present invention. Figure 3 As shown, in step 220, based on the difference between each dimension feature and the reference feature, predicting whether the driver's accelerator operation is correct specifically includes: Step 221, fusing the features of each dimension to obtain a fused feature; Step 222 , based on the difference between the features of each dimension and the baseline features, and the difference between the fusion features and the baseline features, predict whether the driver's accelerator operation is correct.

[0068] Specifically, in order to further improve the accuracy of mis-operation on the accelerator, features of each dimension may be fused to obtain fused features. For example, the features of each dimension can be expressed as: [001 (user number), 1 (male), 9 (driving experience), 35 (age), 0 (no voice prompt for fellow passengers), 0 (correct accelerator)], [001 (user number), 114.3321 (latitude), 30.35216 (longitude), 1 (provincial road), 0 (no obstacles), 1 (speed limit), 60km / h (speed limit threshold), 0 (correct accelerator)], [001 (user number), 55 (current vehicle speed km / h), 60 (average vehicle speed km / h), 50 (speed range left interval value km / h), 70 (speed range right interval value km / h), 1 (braking status), 3 (marked acceleration km / s), 0 (correct accelerator)], [001 (user number), 2 (good weather), 10 (month), 21 (temperature), 15 (time), 98 (visibility), 0 (correct accelerator)].

[0069] The features of the above four dimensions are fused, and the obtained fusion features can be expressed as: [001 (user number), 1 (male), 9 (driving experience), 35 (age), 0 (no voice prompt for fellow passengers), 0 (correct accelerator pedal), 114.3321 (latitude), 30.35216 (longitude), 1 (provincial road), 0 (no obstacles), 1 (speed limit), 60km / h (speed limit threshold), 55 (current vehicle speed km / h), 60 (average vehicle speed km / h), 50 (vehicle speed range left interval value km / h), 70 (vehicle speed range right interval value km / h), 1 (braking status), 3 (marked acceleration km / s), 2 (good weather), 10 (month), 21 (temperature), 15 (time), 78 (visibility), 0 (correct accelerator pedal pedal)].

[0070] Then the difference between the fusion feature and the baseline feature, as well as the difference between the features of each dimension and the baseline feature, is calculated to predict whether the driver's accelerator operation is correct. The difference here can be measured by the chi-square distance.

[0071] In some embodiments, for the prediction module, a BP neural network with 5 nodes in the input layer, 5 nodes in the hidden layer, and 1 node in the output layer can be used for fuzzy calculation. Figure 4 It is a schematic diagram of the structure of the prediction module provided by the present invention. Figure 4 As shown in the figure, the driving factor distance refers to the difference between the characteristics of the driving-related dimension and the benchmark characteristics, the position factor distance refers to the difference between the characteristics of the vehicle position road dimension and the benchmark characteristics, the vehicle factor distance refers to the difference between the characteristics of the vehicle driving dimension and the benchmark characteristics, the external factor distance refers to the difference between the characteristics of the environmental meteorological dimension and the benchmark characteristics, and the comprehensive factor distance refers to the difference between the fusion characteristics and the benchmark characteristics.

[0072] First, the normalized five distances are used as input, and the output is 1 for mis-stepping the accelerator and 0 for correct stepping on the accelerator. Then, these labeled data are used as input again to calculate the output value and determine the probability threshold, which is usually around 0.5. When in use, the five distances of the real-time situation are used as input, the probability of the output is calculated, and compared with the threshold for judgment. If the probability is greater than the threshold, the driver is reminded to mis-step on the accelerator.

[0073] Based on any of the above embodiments, the accelerator operation of the driver is judged, and then the method further includes: If the judgment result indicates that the accelerator pedal operation is an erroneous operation, a reminder message is sent.

[0074] Specifically, considering that the existing solution for suppressing accidental stepping is that the vehicle computer system suppresses acceleration, in special scenarios, there is a possibility of misjudgment by the vehicle computer system. If the vehicle computer system misjudges and erroneously triggers the instruction to suppress acceleration, there is also a safety risk.

[0075] In this embodiment, if the judgment result indicates that the driver's accelerator operation is an error, a reminder message is sent. The reminder message can be a visual signal or a sound signal, such as using a voice prompt to prompt the driver to actively withdraw the accelerator operation, which adds a step of active judgment by the driver and is safer.

[0076] Based on any of the above embodiments, the method for judging the mis-stepping further includes: Identify the driver; If the identity recognition result indicates that the driver is a new user, a mis-stepping judgment model corresponding to the driver is constructed based on the current driving data; If the identity recognition result indicates that the driver is not a new user, a mistaken stepping judgment model corresponding to the driver is loaded, and the mistaken stepping judgment model is updated based on the current driving data.

[0077] Specifically, the driver's facial image or other biometric information is collected in real time using equipment such as on-board cameras and face recognition technology. Other identity authentication methods can also be used, such as fingerprint recognition, iris recognition, or verification of physical credentials such as car keys. The collected identity information is compared with the driver database pre-stored in the system. If the driver's information already exists in the database, an identity match is performed; if not, it is determined to be a new user.

[0078] If the identification result indicates that the driver is a new user, the system needs to build a misoperation judgment model for the driver. During the first driving process, the system records the current driving data in real time. A comprehensive analysis is performed based on the driver's driving habits (such as sudden acceleration, sudden braking, etc.) and driving environment (such as weather, road conditions, etc.) to form the driver's misoperation judgment model.

[0079] If the identification result indicates that the driver is not a new user, the system should update the mis-stepping judgment model corresponding to the driver based on the driver's current driving data. During each driving process, the system will record new driving data in real time and regularly maintain training sets for model training to provide more personalized and accurate predictions.

[0080] Based on any of the above embodiments, Figure 5 FIG. 2 is a flow chart of the method for judging the misstepping provided by the present invention. Figure 5 As shown, the method includes: First, the driver's identity is confirmed through the vehicle cabin sensor, such as camera face recognition. If the identity recognition result indicates that the driver is a new user, a wrong step judgment model corresponding to the driver is constructed; if the identity recognition result indicates that the driver is not a new user, the driver's personalized wrong step judgment model is loaded.

[0081] Secondly, during the driving process, the driver obtains the current driving data in real time, wherein the current driving data includes at least one of the vehicle position road information, vehicle driving information, environmental weather information, and driving related information.

[0082] Finally, based on the current driving data and the wrong-stepping judgment model corresponding to the driver, the driver's accelerator operation is judged. If it is determined that the current operation is the wrong accelerator (according to the model, the brake should be pressed in the current scenario, but the actual driving operation of the driver is the accelerator), then the driver is reminded by voice whether he has accidentally stepped on the accelerator.

[0083] Finally, each piece of scene data can be recorded as effective information, and the training set can be regularly maintained for model training to provide more personalized and accurate predictions.

[0084] The method provided by the embodiment of the present invention combines the big data of the Internet of Vehicles and AI vision and voice technology, recognizes the driver through face recognition, matches the personalized driving models of different drivers, accesses data such as traffic elements, interference elements and other elements visually recognized outside the vehicle, combines data such as automatic positioning and road regulations, compares the difference between the current driving data and the data of the driver's personalized driving model, and determines whether to issue a wrong step prompt information to help the driver correct the driving operation. Based on the data collected by the machine learning algorithm training, the more mileage the driver drives and the more training, the more accurate the prediction will be. Sending voice reminders through predicted data to assist the driver in correcting the driving behavior of wrong stepping adds a step for the driver to actively judge and prevent secondary errors caused by misjudgment.

[0085] The device for judging the wrong stepping provided by the present invention is described below. The device for judging the wrong stepping described below and the method for judging the wrong stepping described above can be referred to each other.

[0086] Based on any of the above embodiments, Figure 6 is a schematic diagram of the structure of the device for judging mis-stepping provided by the present invention. Figure 6 As shown, the device comprises: The data acquisition unit 610 is used to acquire current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information; The mis-operation judgment unit 620 is used to judge the driver's accelerator operation based on the current driving data and a mis-operation judgment model corresponding to the driver, wherein the mis-operation judgment model is trained based on the driver's historical driving data.

[0087] The device for judging the accidental stepping of the accelerator provided in an embodiment of the present invention obtains an accidental stepping judgment model corresponding to the driver by collecting the driver's historical driving data for training. The accidental stepping judgment model can reflect the driver's personal driving habits. Compared with the general configuration parameter thresholds used in related technologies, the accidental stepping judgment model in this embodiment is more personalized and targeted, and can effectively make correct judgments based on different driving habits. Therefore, it can more accurately predict whether the accelerator is accidentally stepped on, thereby improving the accuracy of the accidental stepping judgment.

[0088] Based on any of the above embodiments, it also includes a model acquisition unit, which is used to: Extracting features of each information dimension from the driver's historical driving data to obtain features of each dimension, and determining a baseline feature of the driver based on the features of each dimension; Based on the differences between the features in each dimension and the reference features, the accelerator operation of the driver is predicted, and the mis-accelerator judgment model is trained based on the prediction results.

[0089] Based on any of the above embodiments, the model acquisition unit is specifically used to: Predicting the driver's accelerator operation based on the information gain weight of each dimensional feature and the difference between each dimensional feature and the reference feature; The information gain weight of any dimensional feature is determined based on the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is included, and the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is not included.

[0090] Based on any of the above embodiments, the model acquisition unit is specifically used to: Fusing the features of each dimension to obtain a fused feature; Based on the difference between the features in each dimension and the reference feature, and the difference between the fusion feature and the reference feature, the accelerator operation of the driver is predicted.

[0091] Based on any of the above embodiments, the model acquisition unit is specifically used to: The reference feature is determined based on an average value of features in each dimension corresponding to the case where the driver steps on the accelerator correctly.

[0092] Based on any of the above embodiments, it further includes a reminder unit, which is used to: If the judgment result indicates that the accelerator pedal operation is an erroneous operation, a reminder message is sent.

[0093] Based on any of the above embodiments, it also includes an identity recognition unit, which is used to: Identifying the driver; If the identity recognition result indicates that the driver is a new user, constructing a wrong-stepping judgment model corresponding to the driver based on the current driving data; If the identity recognition result indicates that the driver is not a new user, the mis-stepping judgment model corresponding to the driver is updated based on the current driving data.

[0094] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the method for judging the wrong pedaling, which includes: obtaining current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental meteorological information, and driving related information; judging the accelerator operation of the driver based on the current driving data and the wrong pedaling judgment model corresponding to the driver, wherein the wrong pedaling judgment model is obtained by training based on the driver's historical driving data.

[0095] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0096] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for judging the accelerator pedal incorrectly provided by the above-mentioned methods, and the method includes: obtaining current driving data, and the current driving data includes at least one of vehicle position road information, vehicle driving information, and environmental meteorological information, as well as driving related information; based on the current driving data and a wrong stepping judgment model corresponding to the driver, judging the driver's accelerator pedal operation, and the wrong stepping judgment model is trained based on the driver's historical driving data.

[0097] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the method for judging the accelerator pedal mistake provided by the above-mentioned methods. The method includes: obtaining current driving data, the current driving data including at least one of vehicle position road information, vehicle driving information, environmental meteorological information, and driving related information; judging the accelerator pedal operation of the driver based on the current driving data and a accelerator pedal mistake judgment model corresponding to the driver, the accelerator pedal mistake judgment model being trained based on the driver's historical driving data.

[0098] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for judging an accidental stepping, characterized in that: include: Acquiring current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information; The accelerator pedal operation of the driver is judged based on the current driving data and a wrong pedal judgment model corresponding to the driver, wherein the wrong pedal judgment model is obtained by training based on the driver's historical driving data.

2. The method for judging an accidental stepping on a vehicle according to claim 1, wherein: The step of acquiring the mistaken stepping judgment model comprises: Extracting features of each information dimension from the driver's historical driving data to obtain features of each dimension, and determining a baseline feature of the driver based on the features of each dimension; Based on the differences between the features in each dimension and the reference features, the accelerator operation of the driver is predicted, and the mis-accelerator judgment model is trained based on the prediction results.

3. The method for judging an accidental stepping on a vehicle according to claim 2, wherein: The predicting the accelerator operation of the driver based on the difference between the features of each dimension and the reference features includes: Predicting the driver's accelerator operation based on the information gain weight of each dimensional feature and the difference between each dimensional feature and the reference feature; The information gain weight of any dimensional feature is determined based on the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is included, and the probabilities of the driver pressing the accelerator correctly and incorrectly when the any dimensional feature is not included.

4. The method for judging an accidental stepping on a vehicle according to claim 2, wherein: The predicting the accelerator operation of the driver based on the difference between the features of each dimension and the reference features includes: Fusing the features of each dimension to obtain a fused feature; Based on the difference between the features in each dimension and the reference feature, and the difference between the fusion feature and the reference feature, the accelerator operation of the driver is predicted.

5. The method for judging an accidental stepping on a vehicle according to claim 2, wherein: The determining of the driver's baseline characteristics based on the characteristics of each dimension includes: The reference feature is determined based on an average value of features in each dimension corresponding to the case where the driver steps on the accelerator correctly.

6. The method for judging mis-stepping according to any one of claims 1 to 5, characterized in that: The step of judging the accelerator operation of the driver further includes: If the judgment result indicates that the accelerator pedal operation is an erroneous operation, a reminder message is sent.

7. The method for judging mis-stepping according to any one of claims 1 to 5, characterized in that: The method further comprises: Identifying the driver; If the identity recognition result indicates that the driver is a new user, constructing a wrong-stepping judgment model corresponding to the driver based on the current driving data; If the identity recognition result indicates that the driver is not a new user, a mistaken stepping judgment model corresponding to the driver is loaded, and the mistaken stepping judgment model is updated based on current driving data.

8. A device for judging a wrong step, characterized in that: include: A data acquisition unit, used to acquire current driving data, wherein the current driving data includes at least one of vehicle position road information, vehicle driving information, environmental weather information, and driving related information; The mis-operation judgment unit is used to judge the accelerator operation of the driver based on the current driving data and a mis-operation judgment model corresponding to the driver, wherein the mis-operation judgment model is trained based on the driver's historical driving data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for judging the wrong stepping on the vehicle as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for judging the wrong stepping on the vehicle is implemented as claimed in any one of claims 1 to 7.

Images