CN112758100A - Method and device for mistakenly stepping on accelerator - Google Patents
Method and device for mistakenly stepping on accelerator Download PDFInfo
- Publication number
- CN112758100A CN112758100A CN202110146630.XA CN202110146630A CN112758100A CN 112758100 A CN112758100 A CN 112758100A CN 202110146630 A CN202110146630 A CN 202110146630A CN 112758100 A CN112758100 A CN 112758100A
- Authority
- CN
- China
- Prior art keywords
- data
- accelerator
- acquiring
- driving
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W2050/0001—Details of the control system
- B60W2050/0019—Control system elements or transfer functions
- B60W2050/0028—Mathematical models, e.g. for simulation
- B60W2050/0029—Mathematical model of the driver
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2540/00—Input parameters relating to occupants
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Auxiliary Drives, Propulsion Controls, And Safety Devices (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The invention discloses a detection method for mistakenly stepping on an accelerator and a device using the same, wherein the method comprises the following steps: the method comprises the steps that an acquisition device and a system acquire accelerator treading and related real-time driving data to form a driving information database; according to the known data, carrying out data preprocessing on the data set to obtain characteristic value information of the data set; dividing a target data set into a training set and a test set; optimizing initial parameters by inputting a training set, and establishing an end-to-end optimal training model; and inputting the input test set into the training model to obtain a prediction result, and completing detection. The invention adopts the machine learning and data analysis technology to analyze the driving accelerator mistaken stepping and the related driving data, and has the innovation points that the association rule of the driving data is analyzed through the SVM technology, a detection model with high accuracy and high robustness is established, and the accuracy rate of the accelerator mistaken stepping detection is improved.
Description
Technical Field
The invention relates to a method for stepping on an accelerator by mistake, in particular to a method for detecting misoperation of the accelerator by machine learning and data analysis modeling, and further relates to equipment for stepping on the accelerator by mistake.
Background
Nowadays, the automatic driving technology becomes the latest development direction of the whole automobile industry, and the safety of automobile driving can be comprehensively improved by applying the automatic driving technology. The automatic driving technology realizes semi-automatic driving nowadays, provides driving support for multiple operations in a steering wheel and acceleration and deceleration through a driving environment, and other driving actions are operated by a human driver.
The automatic driving active safety problem is a ring of an automatic driving core, threat analysis is started from the outside of a vehicle and in the vehicle, and an accelerator mistaken stepping detection technology is an important ring for guaranteeing the safety of the automatic driving in the vehicle. Although previous researches have provided a lot of solutions to the problem of mistaken stepping on the accelerator, the problem of mistaken stepping on the accelerator is not analyzed in combination with scene multi-source data for a single scene, and no solution is available for perfectly solving the problem of mistaken stepping on the accelerator.
Several solutions are disclosed in the prior art, including:
the publication number CN106985664A discloses an intelligent automatic control device and a control method for preventing a driver from stepping on an accelerator by mistake, which realize control by adding a human body physiological index measuring and sensing unit, acquiring whether the driver is in a tension state in real time, and judging whether the behavior of the driver belongs to stepping on the accelerator by mistake.
The publication No. CN110466351A discloses a detection control method for preventing the passenger car from being stepped on by mistake, which effectively detects the operation of stepping on by mistake of the accelerator by analyzing the speed, the position information of the accelerator pedal and the position information of the brake pedal, thereby controlling the power system of the whole car to cut off the torque output and controlling the brake system to brake to prevent the passenger car from being stepped on by mistake.
However, none of the prior art disclosures analyzes the driving state by associating various driving data, excavates data association rules, and establishes a highly robust accelerator mistaken-stepping model, thereby improving driving safety.
Disclosure of Invention
In order to overcome the defects of the prior art, the inventor of the invention finds that a high-robustness accelerator mistaken-stepping detection model can be learned through analyzing and processing mass driving associated data and a machine learning method, and the accelerator mistaken-stepping behavior can be accurately identified in application scenes with different speeds. The technical method adopted by the inventor is as follows:
a detection method for mistakenly stepping on an accelerator is characterized by comprising the following steps:
step 1: acquiring experimental data, and acquiring related real-time driving data through data acquisition equipment and a system to form a driving information database;
step 2: analyzing and processing data: according to the known data, carrying out data preprocessing on the data set to obtain characteristic value information of the data set;
and step 3: generating a training set and a testing set: dividing a target data set into a training set and a test set;
and 4, step 4: establishing a model: optimizing initial parameters by inputting a training set, and establishing an end-to-end optimal training model;
and 5: and (4) detecting a result: and inputting the input test set into the training model to obtain a prediction result.
Further, the preprocessing in the step 2 is to perform supplementary processing on the missing values by using an average filling method, then perform feature coding on the data, normalize the feature vector to be between 0 and 1, and then construct a feature vector [ d, x1, x 2.. xn,y]And noise epsilon is added to the eigenvector.
Further, the step 2 of obtaining the feature information of the data set specifically includes: using a point corresponding to a zero mean Gaussian random variable with a known variance and a y coordinate for interference, and recording the generated noise band as xi,x′iWherein the probability density function of ε is:
further, in step 3, the target data set is divided into a training set and a test set, specifically, the data set is divided into K mutually exclusive subsets with similar sizes, each subset maintains independence of data distribution, a union of K-1 subsets is used as the training set each time, and the rest subset is used as the test set, so that K times of training and testing are performed
Further, the value of K is set to logn, and n/K >3d is guaranteed, wherein n represents the data volume and d represents the characteristic number.
Further, the model is specifically established in step 4 by giving the training set T { (x1, y1), (x2, y2), (x 2), andn,yn) In which xnIs the n-th feature vector, ynFor the state flag, i ═ 1, 2.. n, a gaussian kernel function K (x, z) and a penalty function C are selected>0, constructing and solving a convex quadratic optimization problem, and constructing a classification decision function
Further, C is 0.8, and a linear gaussian kernel function is used as the gaussian kernel function; solving an optimization equationGet the optimal solution
Further, the real-time driving data in the step 1 comprises an accelerator pedal signal value.
Further, real-time driving data still includes: one or more of steering wheel rotation parameters, brake pedal data, planned driving trajectory data, traffic environment information, driver information, vehicle weight information.
A device for realizing the detection method of the mistaken stepping on the accelerator comprises an acquisition unit and a processing unit, wherein the acquisition unit is used for acquiring input data and processing requirements; the system comprises a data acquisition device and a system, wherein the data acquisition device is used for acquiring accelerator pedaling and related real-time driving data to form a driving information database, the processing unit is used for carrying out data preprocessing on the acquired data, and inputting the driving data subjected to data preprocessing into a pre-trained model to obtain a detection result.
Further, the acquiring unit may be an accelerator data acquiring device for acquiring an accelerator pedal signal value parameter, a gear lever data acquiring device for acquiring gear lever data, a brake data acquiring device for acquiring brake pedal data, a direction data acquiring device for acquiring a steering wheel rotation parameter, or different combinations of the above acquiring devices.
Further, the acquisition unit is driving state information acquisition device, and further can be including gathering user's travel information's GPRS positioner, big dipper navigation positioner, gather traffic environment information's sensor, like image sensor, sound sensor, gather navigating mate information's sensor, like heart rate sensor, alcohol test sensor, gather the weight sensor of whole car information.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention trains and detects 95% of accelerator misstep operation by collecting the vehicle speed, the accelerator pedal position information and the brake pedal position information.
2. The invention also discovers association rules in the data through a machine learning algorithm by acquiring various driving association data, improves the model precision, is suitable for various emergency extreme conditions, and trains a driving accelerator mistaken-stepping model with high robustness.
The invention can accurately detect and identify the behavior of mistakenly stepping on the accelerator through the release time of the brake pedal, the change rate of the accelerator pedal and the depth of the accelerator pedal in different application scenes, effectively avoids the occurrence of accidents, is suitable for city buses with various driving energy sources such as traditional power, hybrid power and fuel cell electric vehicles, hybrid power vehicles, hydrogen energy power vehicles, pure electric vehicles, other new energy vehicles and the like, and has wide application range.
Drawings
FIG. 1 is a flow chart of model building of the accelerator false stepping detection method of the invention;
FIG. 2 is a signal output data plot of a normal drive accelerator pedal according to the present invention;
FIG. 3 is a signal output data diagram of an accelerator pedal for normal acceleration driving according to the present invention;
FIG. 4 is a signal output data diagram of the driving accelerator pedal of the invention when stepping on the accelerator pedal by mistake;
FIG. 5 is an experimental graph of classification of accelerator pedal data based on SVM of the present invention;
FIG. 6 is a block diagram of the accelerator step error device according to the present invention;
FIG. 7 is an illustration of an acquisition unit device of the accelerator step error device according to the present invention;
FIG. 8 is an illustration of an example of the processing unit of the accelerator step-by-step device according to the present invention.
Detailed Description
The present invention is further illustrated by the following figures and specific examples, which are to be understood as illustrative only and not as limiting the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalent modifications thereof which may occur to those skilled in the art upon reading the present specification.
For the sake of simplicity, the drawings only schematically show the parts relevant to the present invention, and they do not represent the actual structure as a product. In addition, in order to make the drawings concise and understandable, components having the same structure or function in some of the drawings are only schematically illustrated or only labeled. In this document, "one" means not only "only one" but also a case of "more than one".
The first embodiment is as follows:
fig. 1 is a flow chart of model building of the accelerator step-on detection method of the invention, and as shown in fig. 1, the accelerator step-on detection method of the invention comprises:
step 1: and acquiring experimental data, and acquiring related real-time driving data through data acquisition equipment and a system to form a driving information database.
Step 2: analyzing and processing data, performing supplementary processing on missing values by adopting an average filling method according to information in the driving information database in the step 1, then performing feature coding on the data, normalizing feature vectors to be between 0 and 1, and then constructing feature vectors [ d, x1, x 2.. xn,y]And noise epsilon is added to the eigenvector.
And step 3: and generating a training set and a testing set, dividing the data set into K mutually exclusive subsets with similar sizes, wherein each subset keeps the independence of data distribution, a union set of K-1 subsets is used as the training set each time, and the rest subset is used as the testing set, so that K times of training and testing are performed.
And 4, step 4: modeling by giving a training set T { (x1, y1), (x2, y2)n,yn) In which xnIs the n-th feature vector, ynIs a status flag, i ═ 1, 2. The method combines normal constant speed driving data with normal acceleration driving data, and randomly scrambles the data; selecting a Gaussian kernel function K (x, z) and a penalty function C>0, constructing and solving a convex quadratic optimization problem, and constructing a classification decision function
And 5: and (5) result detection, namely inputting the input test set into the training model to obtain a prediction result.
In this embodiment:
in step 1, the experimental data are accelerator pedal voltage value parameters. The accelerator pedal has two paths of signal outputs, the relationship of the two paths of signals is about 2 times, when the two paths of signals are not 2 times, the accelerator pedal is judged to be in fault by the mistaken stepping detection device, and the received signals fail; one path of data is collected, the initial voltage is 0.2V, the maximum voltage is 2V, the data is collected once every 1MS, and the total stroke is 200mm (the opening of the pedal is 0-100%). By analyzing the accelerator pedal signal, the output reaches the accelerator mistaken-stepping remediator, and the accelerator mistaken-stepping remediator reaches the engine computer board.
In step 2, Gaussian smooth noise epsilon is added into the feature vector, the interference is carried out by using a point corresponding to a Gaussian random variable with a known variance and zero mean value and a y coordinate, and the generated noise band is marked as xi,x′iAnd as shown in the second, third and fourth graphs, the distribution after data processing is presented, wherein the probability density function of epsilon is:
in step 3, K is set to be approximately equal to log n, and n/K is guaranteed to be greater than 3d, wherein n represents the data volume and d represents the characteristic number.
In step 4, C is set to be 0.8, and a linear Gaussian kernel function is used as the Gaussian kernel function; solving an optimization equationGet the optimal solution
By the method for detecting the mistaken stepping of the accelerator, the accuracy of the model trained by collecting the information of the accelerator pedal is more than 98.3%.
Although the steps are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed alternately or in alternation with other steps or at least some of the sub-steps or stages of other steps. And the invention does not limit the execution times of the steps, and can be repeatedly executed according to actual requirements, for example, in the process of processing the data in the step 2 to obtain the data characteristic information, if the data acquired in the step 1 has certain defects, the step 1 can be repeatedly executed until the data characteristic obtained in the step 2 passes, and meanwhile, in the process of dividing the training set and the test set, the data amount is considered to be insufficient, or the data has some defects, or the step 1 can be returned to for repeated data acquisition. If certain defects are found in the result detection process, the result detection can be performed again, and according to the above, the steps in the method can be performed in different sequences and different times, so as to achieve the optimal model training effect.
Example two:
on the basis of the first embodiment, in the step 1, the experimental data is composed of an accelerator pedal signal value parameter and a steering wheel rotation parameter, and when the steering wheel rotation parameter and the accelerator pedal signal exceed a set threshold value, the accelerator mistaken-stepping detection device considers that the accelerator pedal is in a fault, and the received signal fails.
Compared with example 1, in this example, after using the steering wheel rotation parameter, the accuracy of the training model is 98.4015%, which is improved by about 0.1% compared with example 1. Meanwhile, when the steering wheel rotation parameters are detected to be obviously different from normal values, driving can be controlled as early as possible, and driving risks are reduced.
Example three:
on the basis of the second embodiment, in the step 1, the experimental data is composed of an accelerator pedal signal value parameter, a steering wheel rotation parameter and brake pedal data, and by analyzing the signals of the accelerator pedal signal, the steering wheel rotation parameter and the brake pedal data, when the steering wheel rotation parameter, the accelerator pedal signal and the brake pedal data exceed the set threshold values, the accelerator mis-stepping detection device considers that the accelerator pedal is in fault, and the received signal fails.
In this example, the accuracy of the training model after further addition of the brake pedal data was 98.4225% compared to example 2, which is an improvement of about 0.01% compared to example 2.
Example four:
on the basis of the first embodiment, in the step 1, the experimental data is composed of accelerator pedal signal value parameters and brake pedal data, and by analyzing the signals of the accelerator pedal signal and the brake pedal data, when the accelerator pedal signal and the brake pedal data exceed the set threshold, the accelerator mistaken-stepping detection device considers that the accelerator pedal is in a fault, and the received signal fails.
Compared with example 1, in this example, after using the steering wheel rotation parameter, the accuracy of the training model is 98.353%, which is improved by about 0.05% compared with example 1.
Example five:
on the basis of the first embodiment, in the step 1, the experimental data is composed of an accelerator pedal signal value parameter and planned driving track data, the driving track data is planned by analyzing a signal pedal accelerator pedal signal, the specified speed limit information of the current driving road section can be read through the line information of the planned driving track data in the map data, and when the accelerator data exceeds the current specified speed limit information, the probability of mistakenly stepping on the accelerator is increased, so that whether the accelerator is mistakenly stepped on or not is judged more accurately.
Example six:
on the basis of the first embodiment, in the step 1, the experimental data is composed of accelerator pedal signal value parameters and traffic environment information, the current driving environment and the surrounding traffic jam condition can be provided by analyzing the signal of the accelerator pedal and the traffic environment information, and by adding the traffic environment information, the real-time traffic environment information can provide the current driving environment.
Example seven:
on the basis of the first embodiment, in the step 1, the experimental data is composed of accelerator pedal signal value parameters and driver information, the driver information can reflect the state of the driver at that time by analyzing the signal of the accelerator pedal, and the driver information is added through the driver information, and through heart rate change and even the condition of alcohol content in the vehicle, when the driver information is seriously abnormal, the probability of mistakenly stepping on the accelerator is increased, so that whether the accelerator is mistakenly stepped on is more accurately judged.
Example eight:
on the basis of the first embodiment, in the step 1, the experimental data is composed of an accelerator pedal signal value parameter and vehicle weight information, the driving state can be analyzed by analyzing the signal pedal accelerator pedal signal and the vehicle weight information and by analyzing the vehicle weight information and the accelerator pedal signal, and when the accelerator is mistakenly stepped on, the vehicle can be stopped in time according to the vehicle weight information, so that the accident loss is reduced.
Example nine:
on the basis of the first embodiment, in the step 1, the experimental data is composed of the accelerator pedal signal value parameter and the gear lever data, the acceleration of the vehicle can be analyzed through the gear lever data and the accelerator pedal signal value parameter, and the acceleration analysis is more accurate in detection of mistaken stepping of the accelerator compared with direct speed analysis.
The invention also relates to an accelerator false stepping detection device for realizing the accelerator false stepping detection method, as shown in fig. 6-8, the accelerator false stepping detection device comprises an acquisition unit 610 and a processing unit 620, wherein the acquisition unit 610 is used for acquiring input data and processing requirements; the accelerator pedal and related real-time driving data are collected through the data collection equipment and the data collection system to form a driving information database.
And the processing unit 620 is used for preprocessing the acquired data and inputting the driving data subjected to data preprocessing into the pre-trained model to obtain a detection result.
The obtaining unit 610 may be an accelerator data collecting device for obtaining an accelerator pedal signal value parameter, a gear lever data collecting device for collecting gear lever data, a brake data collecting device for collecting brake pedal data, a direction data collecting device for collecting a steering wheel rotation parameter, or different combinations of the above-mentioned collecting devices.
The acquisition unit 610 may also be a driving state information acquisition device, and further may include a GPRS positioning device for acquiring user travel information, a beidou navigation positioning device, a sensor for acquiring traffic environment information, such as an image sensor and a sound sensor, a sensor for acquiring information of a driver, such as a heart rate sensor, an alcohol test sensor, a weight sensor for acquiring information of a whole vehicle, and the like.
The processing unit comprises a CPU, a GPU, an FPGA, an AI chip and the like.
The above description is only a preferred embodiment of the present invention, and should not be construed as limiting the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (12)
1. A detection method for mistakenly stepping on an accelerator is characterized by comprising the following steps:
step 1: acquiring experimental data, and acquiring related real-time driving data through data acquisition equipment and a system to form a driving information database;
step 2: analyzing and processing data: according to the known data, carrying out data preprocessing on the data set to obtain characteristic value information of the data set;
and step 3: generating a training set and a testing set: dividing a target data set into a training set and a test set;
and 4, step 4: establishing a model: optimizing initial parameters by inputting a training set, and establishing an end-to-end optimal training model;
and 5: and (4) detecting a result: and inputting the input test set into the training model to obtain a prediction result.
2. The accelerator step-on mistake detecting method according to claim 1, characterized in that: the preprocessing in the step 2 is to perform supplementary processing on the missing value by adopting an average filling method, then perform feature coding on the data, normalize the feature vector to be between 0 and 1, and then construct a feature vector [ d, x1, x 2.. xn,y]And noise epsilon is added to the eigenvector.
3. The accelerator step-on mistake detecting method according to claim 2, characterized in that: the step 2 of obtaining the feature information of the data set specifically includes: using a point corresponding to a zero mean Gaussian random variable with a known variance and a y coordinate for interference, and recording the generated noise band as xi,x′iWherein the probability density function of ε is:
4. the accelerator step-on mistake detecting method according to claim 1, characterized in that: in the step 3, the target data set is divided into a training set and a test set, specifically, the data set is divided into K mutually exclusive subsets with similar sizes, each subset keeps independence of data distribution, a union set of K-1 subsets is used as the training set each time, and the rest subset is used as the test set, so that K times of training and testing are performed.
5. The accelerator false stepping detection method according to claim 4, characterized in that: the value of K is set to log n, and n/K >3d is guaranteed, where n represents the data volume and d represents the feature number.
6. The accelerator step-on mistake detecting method according to claim 1, characterized in that: the modeling in step 4 is specifically performed by giving a training set T { (x1, y1), (x2, y2), (x 1), andn,yn) In which xnIs the n-th feature vector, ynFor the state flag, i ═ 1, 2.. n, a gaussian kernel function K (x, z) and a penalty function C are selected>0, constructing and solving a convex quadratic optimization problem, and constructing a classification decision function
8. The accelerator step-on mistake detecting method according to claim 1, characterized in that: the real-time driving data in the step 1 comprises an accelerator pedal signal value.
9. The accelerator false stepping detection method according to claim 8, characterized in that: the real-time driving data further comprises: one or more of steering wheel rotation parameters, brake pedal data, planned driving trajectory data, traffic environment information, driver information, vehicle weight information.
10. A device for realizing the accelerator false stepping detection method of claim 1 comprises an acquisition unit and a processing unit, and is characterized in that: the acquisition unit is used for acquiring input data and processing requirements; the system comprises a data acquisition device and a system, wherein the data acquisition device is used for acquiring accelerator pedaling and related real-time driving data to form a driving information database, the processing unit is used for carrying out data preprocessing on the acquired data, and inputting the driving data subjected to data preprocessing into a pre-trained model to obtain a detection result.
11. The device for realizing the accelerator step-on mistake detecting method according to claim 10, characterized in that: the acquiring unit may be an accelerator data acquiring device for acquiring an accelerator pedal signal value parameter, a gear lever data acquiring device for acquiring gear lever data, a brake data acquiring device for acquiring brake pedal data, a direction data acquiring device for acquiring a steering wheel rotation parameter, or different combinations of the above acquiring devices.
12. The device for realizing the accelerator step-on mistake detecting method according to claim 10, characterized in that: the acquisition unit is driving state information acquisition device, and further can be including gathering user's travel information's GPRS positioner, big dipper navigation positioner, the sensor of gathering traffic environment information, like image sensor, sound sensor, gather navigating mate information's sensor, like heart rate sensor, alcohol test sensor, gather the weight sensor of whole car information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110146630.XA CN112758100B (en) | 2021-02-03 | 2021-02-03 | Accelerator mistaken stepping detection method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110146630.XA CN112758100B (en) | 2021-02-03 | 2021-02-03 | Accelerator mistaken stepping detection method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112758100A true CN112758100A (en) | 2021-05-07 |
CN112758100B CN112758100B (en) | 2023-03-14 |
Family
ID=75704769
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110146630.XA Active CN112758100B (en) | 2021-02-03 | 2021-02-03 | Accelerator mistaken stepping detection method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112758100B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102445902A (en) * | 2010-09-30 | 2012-05-09 | 西门子公司 | System and method for conditional multi-output regression for machine condition monitoring |
CN107679557A (en) * | 2017-09-19 | 2018-02-09 | 平安科技(深圳)有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
WO2019056499A1 (en) * | 2017-09-20 | 2019-03-28 | 平安科技(深圳)有限公司 | Prediction model training method, data monitoring method, apparatuses, device and medium |
CN110745136A (en) * | 2019-09-20 | 2020-02-04 | 中国科学技术大学 | Driving self-adaptive control method |
CN111401565A (en) * | 2020-02-11 | 2020-07-10 | 西安电子科技大学 | DOA estimation method based on machine learning algorithm XGboost |
CN111582534A (en) * | 2020-03-09 | 2020-08-25 | 华能四川水电有限公司 | Hydroelectric power generation prediction method based on extreme learning machine |
CN112183591A (en) * | 2020-09-15 | 2021-01-05 | 上海电机学院 | Transformer fault diagnosis method based on stack sparse noise reduction automatic coding network |
-
2021
- 2021-02-03 CN CN202110146630.XA patent/CN112758100B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102445902A (en) * | 2010-09-30 | 2012-05-09 | 西门子公司 | System and method for conditional multi-output regression for machine condition monitoring |
CN107679557A (en) * | 2017-09-19 | 2018-02-09 | 平安科技(深圳)有限公司 | Driving model training method, driver's recognition methods, device, equipment and medium |
WO2019056499A1 (en) * | 2017-09-20 | 2019-03-28 | 平安科技(深圳)有限公司 | Prediction model training method, data monitoring method, apparatuses, device and medium |
CN110745136A (en) * | 2019-09-20 | 2020-02-04 | 中国科学技术大学 | Driving self-adaptive control method |
CN111401565A (en) * | 2020-02-11 | 2020-07-10 | 西安电子科技大学 | DOA estimation method based on machine learning algorithm XGboost |
CN111582534A (en) * | 2020-03-09 | 2020-08-25 | 华能四川水电有限公司 | Hydroelectric power generation prediction method based on extreme learning machine |
CN112183591A (en) * | 2020-09-15 | 2021-01-05 | 上海电机学院 | Transformer fault diagnosis method based on stack sparse noise reduction automatic coding network |
Also Published As
Publication number | Publication date |
---|---|
CN112758100B (en) | 2023-03-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gadepally et al. | A framework for estimating driver decisions near intersections | |
CN108248609A (en) | Hybrid vehicle and the method that prediction drives pattern in hybrid vehicle | |
CN106971194A (en) | A kind of driving intention recognition methods based on the double-deck algorithms of improvement HMM and SVM | |
CN106956680A (en) | A kind of electric automobile driving behavior identifying and analyzing method | |
CN106960189A (en) | A kind of driving intention decision method based on hidden Markov model | |
JP2022089806A (en) | Modeling operation profiles of vehicle | |
Kanaan et al. | Using naturalistic vehicle-based data to predict distraction and environmental demand | |
CN112793576A (en) | Lane change decision method and system based on rule and machine learning fusion | |
Camlica et al. | Feature abstraction for driver behaviour detection with stacked sparse auto-encoders | |
CN113642114B (en) | Personified random following driving behavior modeling method capable of making mistakes | |
CN111340074A (en) | Driver braking intention identification method based on improved RBF neural network | |
CN114492043A (en) | Personalized driver following modeling method considering perception limited characteristics | |
CN112758100B (en) | Accelerator mistaken stepping detection method and device | |
Wei et al. | Least square-support vector regression based car-following model with sparse sample selection | |
Harkous et al. | Application of hidden Markov model on car sensors for detecting drunk drivers | |
CN105869412A (en) | Method for identifying fast acceleration behaviors based on vehicle running data | |
CN108268678A (en) | Driving behavior analysis method, apparatus and system | |
Wang et al. | A data-driven estimation of driving style using deep clustering | |
CN115688003A (en) | Driver identification method and device, computer equipment and readable storage medium | |
Peng et al. | A Method for Vehicle Collision Risk Assessment through Inferring Driver's Braking Actions in Near-Crash Situations | |
CN114148349A (en) | Vehicle personalized following control method based on generation countermeasure simulation learning | |
Zhong et al. | Evaluating the Efficacy of Real-Time Connected Vehicle Basic Safety Messages in Mitigating Aberrant Driving Behaviour and Risk of Vehicle Crashes: Preliminary Insights from Highway Scenarios | |
Wu et al. | Modeling Lead-Vehicle Kinematics for Rear-End Crash Scenario Generation | |
CN115035718B (en) | Scene complexity evaluation method based on vehicle action types and times | |
Barua et al. | Multi-scale Data Fusion and Machine Learning for Vehicle Manoeuvre Classification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |