CN110615001B - Driving safety reminding method, device and medium based on CAN data - Google Patents
Driving safety reminding method, device and medium based on CAN data Download PDFInfo
- Publication number
- CN110615001B CN110615001B CN201910926150.8A CN201910926150A CN110615001B CN 110615001 B CN110615001 B CN 110615001B CN 201910926150 A CN201910926150 A CN 201910926150A CN 110615001 B CN110615001 B CN 110615001B
- Authority
- CN
- China
- Prior art keywords
- driving behavior
- irregular
- data
- value
- vehicle
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000001788 irregular Effects 0.000 claims abstract description 156
- 238000003064 k means clustering Methods 0.000 claims abstract description 14
- 230000006399 behavior Effects 0.000 claims description 232
- 230000001133 acceleration Effects 0.000 claims description 74
- 230000002159 abnormal effect Effects 0.000 claims description 50
- 238000011156 evaluation Methods 0.000 claims description 16
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 8
- 206010027146 Melanoderma Diseases 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000007935 neutral effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000000630 rising effect Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 6
- 238000012546 transfer Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000002131 composite material Substances 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 230000007704 transition Effects 0.000 description 4
- 230000003203 everyday effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
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
- B60W40/09—Driving style or behaviour
-
- 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
- B60W50/08—Interaction between the driver and the control system
- B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
-
- 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
- B60W2050/0062—Adapting control system settings
- B60W2050/0075—Automatic parameter input, automatic initialising or calibrating means
-
- 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
- B60W2556/00—Input parameters relating to data
- B60W2556/10—Historical data
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Time Recorders, Dirve Recorders, Access Control (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a driving safety reminding method, a driving safety reminding device and a driving safety reminding medium based on CAN data. The method is characterized in that a threshold judgment rule is defined to automatically generate an irregular driving behavior rule threshold based on the collected CAN data of the whole vehicles of the fleet and the division of line space and time, the irregular driving behavior rule threshold is updated to vehicle-mounted equipment in real time, when the irregular driving behavior rule threshold is exceeded, a voice prompt is performed on a driver, a Hidden Markov Model (HMM) and a k-means clustering method are further adopted to filter misjudgments in the irregular driving behavior, the operation trends of the driver and each line are statistically analyzed, and a warning is given out.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a driving safety reminding method and device based on CAN data and a storage medium.
Background
Currently, the driving behavior habits of drivers are closely related to the potential safety hazards, the fuel consumption and the service life of vehicles. The oil consumption of the driver with good driving habits is lower than that of other drivers, and meanwhile, the probability of potential safety hazards is also reduced; because of good driving habits, the service life of the vehicle can be correspondingly prolonged. Therefore, it is very necessary to monitor and regulate the driving behavior of the driver.
In the prior art, generally, the method is based on a fixed rule algorithm, cannot automatically adjust according to road condition changes or external environment changes, cannot perform statistical analysis on the bus route and the driver on irregular driving behaviors, and does not establish a specific method for evaluating driving.
Disclosure of Invention
The present invention provides the following technical solutions to overcome the above-mentioned drawbacks in the prior art.
A driving safety reminding method based on CAN data comprises the following steps:
acquiring historical data, namely acquiring state information and black spot data of a collected bus when the bus runs by connecting vehicle-mounted equipment of a bus through a CAN bus, wherein the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine rotating speed, gear signals, horn signals and door signals, and transmitting the acquired state information to a cloud; the black point data comprises the type, longitude and latitude and black point grade of the black point;
the method comprises the following steps of time and space division, wherein a bus running line and time are divided, the line space is divided according to stations and station intervals, black points are introduced for subdivision, time periods are divided according to working days and weekends every week, and then each day is divided into an early peak, a daytime non-peak section, a late peak and a night non-peak section;
and an irregular driving behavior defining step, wherein the irregular driving behavior is defined as follows: the method comprises the following steps of emergent acceleration, emergent deceleration, emergent stop, emergent acceleration for starting, horn alarm, stable door opening of a vehicle when the vehicle is not stopped, vehicle door opening before starting, neutral sliding, large accelerator alarm, cold start idling, irregular arrival, irregular departure and overspeed, wherein the behavior judgment logics of the large accelerator alarm and the cold start idling are fixed;
the method comprises the following steps of (1) determining an abnormal driving behavior threshold value, namely judging whether black points exist in a line space of the abnormal driving behavior except for large throttle alarm and cold start idling, if the line space is a black point and has a hard requirement on the driving behavior, directly using a black point limit value as the threshold value of the abnormal driving behavior, and otherwise, setting basic threshold values of the abnormal driving behavior in different areas and time periods;
and a reminding step, namely when the real-time data in the state information of the running vehicle in the driving process is judged to be the non-standard behavior according to the non-standard driving behavior threshold value, sending a corresponding voice prompt to the driver through the vehicle-mounted equipment.
Further, the setting of the basic threshold of each irregular driving behavior in different areas and time periods is to screen data exceeding or falling below the basic value from historical data, calculate the duration of the behavior for some types of irregular driving behaviors, and then calculate the basic threshold by using a box plot outlier method.
Still further, the method comprises: and an irregular driving behavior threshold value updating step, namely recalculating all threshold values of irregular driving behaviors except for large throttle alarm and cold start idling by using an irregular driving behavior threshold value determining step after the historical data is updated, downloading all updated threshold values to the vehicle-mounted equipment, or judging the region and the time period to which the threshold values belong according to real-time line space and time information when the bus line changes, and automatically updating the threshold values by the vehicle-mounted equipment.
Still further, the method further comprises: and analyzing the driver behavior, wherein the step of analyzing the driver behavior is to count that 80% of data sequence of the driver in the state information of each line space and each time period is a habit value, make a value exceeding the habit value in a certain proportion be an irregular driving behavior critical value, make the data exceeding the critical value be irregular driving behavior data, and filter misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering method to obtain the filtered irregular driving behavior.
Still further, the method further comprises: a driving behavior abnormal value calculation step of calculating an abnormal value α ═ α of the corresponding driving behavior for each of the filtered irregular driving behaviors1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of some index data in the abnormal data, wTIs an abnormal driving behavior critical value.
Still further, the method further comprises: the method comprises the steps of bus route and driver evaluation, wherein the irregular behaviors of bus route spaces in a period are counted, an irregular behavior trend graph of each bus route space is generated, and the lines with the irregular behavior quantity exceeding a certain threshold value or the irregular behaviors rising in proportion exceeding a certain percentage are reminded; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
The invention also provides a driving safety reminding device based on the CAN data, which comprises:
the system comprises a historical data acquisition unit, a cloud terminal and a cloud terminal, wherein the historical data acquisition unit is connected with vehicle-mounted equipment of a bus through a CAN bus to acquire and collect state information and black spot data of the bus during running, the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine rotating speed, gear signals, horn signals and vehicle door signals, and the acquired state information is transmitted to the cloud terminal; the black point data comprises the type, longitude and latitude and black point grade of the black point;
the time-space division unit is used for dividing the bus running line and time, dividing the line space into lines according to stations and station intervals, introducing black points for subdivision, dividing time periods into time periods according to working days and weekends every week, and dividing each day into early peak, daytime non-peak sections, late peak and night non-peak sections;
the irregular driving behavior defining unit defines irregular driving behaviors as follows: the method comprises the following steps of emergent acceleration, emergent deceleration, emergent stop, emergent acceleration for starting, horn alarm, stable door opening of a vehicle when the vehicle is not stopped, vehicle door opening before starting, neutral sliding, large accelerator alarm, cold start idling, irregular arrival, irregular departure and overspeed, wherein the behavior judgment logics of the large accelerator alarm and the cold start idling are fixed;
the abnormal driving behavior threshold value determining unit is used for judging whether black points exist in the line space of the abnormal driving behavior except for large throttle alarm and cold start idling, directly using a black point limit value as a threshold value of the abnormal driving behavior if the line space is the black points and has hard requirements on the driving behavior, and otherwise setting basic threshold values of the abnormal driving behavior in different areas and time periods;
and the reminding unit is used for sending corresponding voice prompt to the driver through the vehicle-mounted equipment when the real-time data in the state information of the running vehicle in the driving process is judged to be the non-standard behavior according to the non-standard driving behavior threshold value.
Further, the setting of the basic threshold of each irregular driving behavior in different areas and time periods is to screen data exceeding or falling below the basic value from historical data, calculate the duration of the behavior for some types of irregular driving behaviors, and then calculate the basic threshold by using a box plot outlier method.
Still further, the apparatus comprises: and the irregular driving behavior threshold value updating unit is used for recalculating each threshold value of irregular driving behaviors except for large throttle alarm and cold start idling by using the irregular driving behavior threshold value determining unit after the historical data is updated, downloading each updated threshold value to the vehicle-mounted equipment, or judging the region and the time period to which the threshold value belongs according to real-time line space and time information when the bus line changes, and automatically updating the threshold value by the vehicle-mounted equipment.
Still further, the apparatus further comprises: and the driver behavior analysis unit is used for counting that 80% of data sequence of the driver in the state information of each line space and each time period is a habit value, determining that a value exceeding the habit value by a certain proportion is an irregular driving behavior critical value, determining that the data exceeding the critical value is irregular driving behavior data, and filtering misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering device to obtain the filtered irregular driving behavior.
Still further, the apparatus further comprises: a driving behavior abnormal value calculation unit for calculating abnormal value alpha of corresponding driving behavior for each filtered irregular driving behavior1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of some index data in the abnormal data, wTIs an abnormal driving behavior critical value.
Still further, the apparatus further comprises: the bus route and driver evaluation unit counts the irregular behaviors of the bus route space in a period and generates an irregular behavior trend chart of each bus route space,
reminding lines of which the number of the irregular behaviors exceeds a certain threshold value or the irregular behaviors rise by a certain percentage in the same ratio; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
The invention also proposes a computer-readable storage medium having stored thereon computer program code which, when executed by a computer, performs any of the methods described above.
The invention has the technical effects that: the method is characterized in that a threshold judgment rule is defined to automatically generate an irregular driving behavior rule threshold based on the collected CAN data of the whole vehicles of the fleet and the division of line space and time, the irregular driving behavior rule threshold is updated to vehicle-mounted equipment in real time, when the irregular driving behavior rule threshold is exceeded, a voice prompt is performed on a driver, a Hidden Markov Model (HMM) and a k-means clustering method are further adopted to filter misjudgments in the irregular driving behavior, the operation trends of the driver and each line are statistically analyzed, and a warning is given out.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings.
Fig. 1 is a flowchart of a driving safety reminding method based on CAN data according to an embodiment of the present invention.
Fig. 2 is a structural diagram of a driving safety warning apparatus based on CAN data according to an embodiment of the present invention.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 shows a driving safety reminding method based on CAN data, which includes:
a historical data acquisition step S101, wherein the method comprises the steps of acquiring state information and black spot data of a bus when the bus runs by connecting vehicle-mounted equipment of the bus through a CAN bus, wherein the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine rotating speed, gear signals, horn signals and vehicle door signals, and transmitting the acquired state information to a cloud end; the black point data comprises the type, longitude and latitude and black point grade of the black point. The collected state information and the collected black point data can be uploaded to the cloud, and processing and analysis can be conveniently carried out by using cloud service.
A space-time division step S102, wherein the bus running line and time are divided, the line space is divided according to the station and the station interval, black points are introduced for subdivision, the time period is divided according to the working day and weekend every week, and then each day is divided into an early peak, a daytime non-peak section, a late peak and a night non-peak section; if a black point exists in the site interval, dividing the site interval according to the site interval 1, the black point and the site interval 2; and if two black points exist in the site interval, dividing the site interval according to the site interval 1-the black point 1-the site interval 2-the black point 2-the site interval 3.
The irregular driving behavior defining step S103 defines the irregular driving behavior as follows: the method comprises the following steps of sudden acceleration, sudden deceleration, sudden stop, starting sudden acceleration, horn alarm, vehicle door opening without stopping stably, vehicle door closing without starting, neutral sliding, large accelerator alarm, cold start idling, irregular entering, irregular exiting and overspeed, wherein behavior judgment logics of the large accelerator alarm and the cold start idling are fixed. For the bus, the judgment rule of large accelerator alarm is that the rotating speed of the engine lasts for more than 2min and exceeds 1500rpm/min, the large accelerator is operated at the moment, the judgment rule of cold start idling is that the rotating speed of the engine cannot exceed 650rpm/min within 5min after the engine is started for the first time, and otherwise, the cold start idling operation is performed. These two logic rules are fixed and do not require machine learning updates to them.
And an irregular driving behavior threshold value determining step S104, for irregular driving behaviors except for large throttle alarm and cold start idling, judging whether black points exist in the line space, if the line space is a black point and has hard requirements on the driving behaviors, directly using a black point limit value as a threshold value of the irregular driving behaviors, and otherwise, setting basic threshold values of the irregular driving behaviors in different areas and time periods. Because the data is in a continuously updated state, the determination of the irregular driving behavior rule threshold value can be automatically issued to the vehicle-mounted intelligent equipment after the reasonable rule threshold value is determined along with different dynamic changes of geographic positions and time periods. The rules are established as follows:
and a reminding step S105, when the real-time data in the state information of the running vehicle in the driving process is judged to be the irregular behaviors according to the irregular driving behavior threshold value, sending corresponding voice prompt to the driver through the vehicle-mounted equipment. For example, the vehicle speed control system can prompt the user to 'not need to be too fast to accelerate' when accelerating suddenly, and can prompt the user to 'control the vehicle speed, not too fast and pay attention to the highest speed limit of each area of the route' when accelerating lightly.
In one embodiment, the setting of the basic threshold of each irregular driving behavior in different areas and time periods is to screen the historical data for data exceeding or falling below the basic value (to select exceeding or falling below, according to the type of the driving behavior), calculate the duration of the behavior for part of the irregular driving behavior types, and then use a box plot outlier method to take the value exceeding a certain proportion as the basic threshold. Taking rapid acceleration as an example, setting basic acceleration of each time period of each area of a line, screening out data of which the acceleration is greater than the basic acceleration, calculating an acceleration basic threshold and a duration basic threshold of different areas and time periods by using a box-type diagram outlier method, and judging that the vehicle is rapid acceleration when the acceleration and the duration exceed the acceleration and the duration basic thresholds. The boxplot outlier algorithm is as follows: data are arranged from large to small, an upper quartile Q3, a middle quartile Q1 and a lower quartile Q1 are respectively calculated, two line segments which are the same as the middle line are drawn at Q3+1.5IQR (quartile spacing, IQR is Q3-Q1) and Q1-1.5IQR, the two line segments are outlier cut-off points and are also called upper/lower edges, and the values of the upper edge or the lower edge are used as basic thresholds.
In one embodiment, the method further comprises: and an irregular driving behavior threshold updating step S106, which is to recalculate each threshold of irregular driving behaviors except for large throttle alarm and cold start idling by using an irregular driving behavior threshold determining step after the historical data is updated, download each updated threshold to the vehicle-mounted equipment, or judge the belonged area and time period according to real-time line space and time information when the bus line is changed, and automatically update the threshold by the vehicle-mounted equipment. When the bus route changes, such as the route is temporarily adjusted, or the route A supports the route B, the area and the time period of the bus can be judged according to real-time route space and time information, the threshold value is automatically updated by the vehicle-mounted equipment, the reminding error is prevented, and the safety is improved. This is one of the important points of the present invention.
In one embodiment, the method further comprises: and a driver behavior analysis step S107, counting that 80% of data sequence of the state information of the driver in each route space and each time period is a habit value, determining that a value exceeding the habit value by a certain proportion is an irregular driving behavior critical value, determining that the data exceeding the critical value is irregular driving behavior data, and filtering misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering method to obtain the filtered irregular driving behavior. By the method, the misjudgments in the irregular driving behaviors can be filtered, so that the accuracy of evaluation on the driver is improved.
The method comprises the steps of counting a value that 80% of data sequence in state information of drivers in each line space and each time period is a habit value, determining a value exceeding the habit value by a certain proportion to be an irregular driving behavior critical value, and determining data exceeding the critical value to be irregular driving behavior data. Taking rapid acceleration as an example, values of the acceleration and duration of the driver in each region and each time period are respectively calculated, wherein the ranking of the acceleration and duration of each time period is 80%, the acceleration and duration exceeding the values by 20% in the region and the time period are judged to be rapid acceleration critical values, and data exceeding the critical values are irregular driving behavior data. Then, firstly determining a driving scene and then clustering (an HMM classifier can only filter misjudged driver behaviors within a certain intention range), wherein a Hidden Markov Model (HMM) is used for filtering the misjudged driver behaviors: the driving intention corresponds to a hidden state of the hidden markov chain, and the driver behavior is regarded as a composite of the driving operation.
The hidden markov model can be divided into a data acquisition layer, a driving operation classification layer, a driving operation sequence transfer layer and a driver intention transfer layer, wherein the driver intention is a driver stable awareness for a long period of time and belongs to a hidden state in the HMM. The driving operation classification layer is a result of data classification, forms driving operation nodes, describes a certain state and basic behaviors of the current vehicle, and each node represents a certain characteristic of driver operation, so that the driving operation classification layer is convenient to visually understand. And a driving operation sequence transfer layer, wherein the driving operation sequence is a composite result of driving operation and is a comprehensive description of the operation of the driver in a short time to form an observation sequence of the model.
And dividing the observation data of the duration according to time periods and regions, training an HMM classifier by adopting a Baum-Welch algorithm, and randomly assigning values to the hidden state transition probability matrix, the observation state transition probability matrix and the initial state probability matrix.
For example, the rapid acceleration operation is an irregular driving behavior determined in a short time, and therefore it is only possible to determine the driver's intention as a part of the result of the driver's intention execution, and it is not possible to directly determine the driver's intention. And if the sudden deceleration behavior has the consciousness of emergency obstacle avoidance, selecting the vehicle speed, the acceleration and the steering as characteristic vectors, intercepting a part of the continuous observation sequence from a sudden deceleration operation point forwards and backwards, and analyzing the characteristics of the observation sequence. Similar denormal driver behavior clustering: and performing k-means clustering on the irregular driving behaviors in time and space, finding out special reasons in a driving scene, and determining the k value of the irregular driving behavior clustering based on space and time by adopting an enumeration and contour coefficient evaluation method. When calculating the unnormal score, the deduction weight may be reduced for some special scenes. This is another important inventive point of the present invention.
In one embodiment, the method further comprises: a driving behavior abnormal value calculation step S108 of calculating an abnormal value α ═ α of the corresponding driving behavior for each of the filtered irregular driving behaviors1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of some index data in the abnormal data, wTIs an abnormal driving behavior critical value.
Setting basic acceleration of each time period of each region of a line by taking rapid acceleration as an example, screening out data of which the acceleration is greater than the basic acceleration, calculating acceleration threshold values and duration threshold values of different regions and time periods by using a box-type diagram outlier method, judging that the vehicle acceleration and the duration exceeding the threshold values are the rapid acceleration, and calculating a global relative deviation value of the rapid acceleration data; respectively calculating values of the driver at 80% of the sequence of the acceleration and the duration of each region and each time period, judging that the acceleration and the duration which exceed the values by 20% in the region and the time period are urgent acceleration critical values, filtering misjudged driver behaviors by an HMM classifier, and then rejecting data of certain scenes by k-means clustering to obtain personal relative deviation values of urgent acceleration; and adding the two relative deviation values to obtain an abnormal value of the driving behavior. This is another important inventive point of the present invention.
In one embodiment, the method further comprises: a bus route and driver evaluation step S109, counting the irregular behaviors of the bus route space in a period, generating an irregular behavior trend graph of each bus route space, and reminding the routes of which the number of the irregular behaviors exceeds a certain threshold value or the irregular behaviors increase more than a certain percentage in a same ratio; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
The method specifically comprises the following steps: counting the irregular behaviors of each bus route space every day/week, warning the route areas where the irregular behaviors are frequently generated, generating an irregular behavior trend graph of each bus route space, and warning the irregular behaviors to a certain rising degree; and counting all driving behavior abnormal values of each driver on average in each working day, and converting the abnormal values into scores of 60-100 according to the ratio after taking the opposite numbers. And (3) scoring the driving behavior of the driver: and counting the average driving behavior abnormal value (weight problem of different driving behaviors) of the driver in each working day, and converting the abnormal value into a score of 60-100 according to a ratio after taking an inverse number. And ranking the driving behavior of each driver from high to low, and selecting the driver with the score of 90 points or more as an excellent driver, the driver with the score of [70,90) as a good driver and the driver with the score of [60,70) as a driver to be improved.
The method is based on the collected CAN data of the running of the total vehicles of all vehicles of a fleet, the irregular driving behavior rule threshold value is automatically generated by using a machine self-learning algorithm after the line space and time are divided, the irregular driving behavior rule threshold value is updated to the vehicle-mounted equipment in real time, when the irregular driving behavior rule threshold value of the line space and time is exceeded, the driver is prompted by voice, the running trends of the driver and each line are statistically analyzed, and the misjudgments in the irregular driving behavior are filtered by adopting a Hidden Markov Model (HMM) and a k-means clustering method, so that the driving behavior of the driver under the emergency condition is prevented from being misrated as the bad driving behavior.
Fig. 2 shows a driving safety reminding device based on CAN data according to the present invention, which comprises:
the historical data acquisition unit 201 is connected with vehicle-mounted equipment of a bus through a CAN bus to acquire state information and black spot data of the bus during running, wherein the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine speed, gear signals, horn signals and door signals, and transmits the acquired state information to a cloud; the black point data comprises the type, longitude and latitude and black point grade of the black point. The collected state information and the collected black point data can be uploaded to the cloud, and processing and analysis can be conveniently carried out by using cloud service.
The space-time division unit 202 is used for dividing the bus running line and time, dividing the line space into lines according to stations and station intervals, introducing black points for subdivision, dividing time periods into time periods according to working days and weekends every week, and dividing each day into early peak, daytime non-peak sections, late peak and night non-peak sections; if a black point exists in the site interval, dividing the site interval according to the site interval 1, the black point and the site interval 2; and if two black points exist in the site interval, dividing the site interval according to the site interval 1-the black point 1-the site interval 2-the black point 2-the site interval 3.
The irregular driving behavior definition unit 203 defines the irregular driving behavior as follows: the method comprises the following steps of sudden acceleration, sudden deceleration, sudden stop, starting sudden acceleration, horn alarm, vehicle door opening without stopping stably, vehicle door closing without starting, neutral sliding, large accelerator alarm, cold start idling, irregular entering, irregular exiting and overspeed, wherein behavior judgment logics of the large accelerator alarm and the cold start idling are fixed. For the bus, the judgment rule of large accelerator alarm is that the rotating speed of the engine lasts for more than 2min and exceeds 1500rpm/min, the large accelerator is operated at the moment, the judgment rule of cold start idling is that the rotating speed of the engine cannot exceed 650rpm/min within 5min after the engine is started for the first time, and otherwise, the cold start idling operation is performed. These two logic rules are fixed and do not require machine learning updates to them.
The irregular driving behavior threshold value determining unit 204 determines whether a black point exists in a line space to which the driver belongs for irregular driving behaviors except for large throttle alarm and cold start idling, directly uses a black point limit value as a threshold value of the irregular driving behavior if the line space is a black point and has a hard requirement on the driving behavior, and otherwise sets a basic threshold value of each irregular driving behavior in different areas and time periods. Because the data is in a continuously updated state, the determination of the irregular driving behavior rule threshold value can be automatically issued to the vehicle-mounted intelligent equipment after the reasonable rule threshold value is determined along with different dynamic changes of geographic positions and time periods. The rules are established as follows:
and the reminding unit 205 is used for sending a corresponding voice prompt to the driver through the vehicle-mounted equipment when the real-time data in the state information of the running vehicle in the driving process is judged to be the irregular behavior according to the irregular driving behavior threshold value. For example, the vehicle speed control system can prompt the user to 'not need to be too fast to accelerate' when accelerating suddenly, and can prompt the user to 'control the vehicle speed, not too fast and pay attention to the highest speed limit of each area of the route' when accelerating lightly.
In one embodiment, the setting of the basic threshold of each irregular driving behavior in different areas and time periods is to screen the historical data for data exceeding or falling below the basic value (to select exceeding or falling below, according to the type of the driving behavior), calculate the duration of the behavior for part of the irregular driving behavior types, and then use a box plot outlier method to take the value exceeding a certain proportion as the basic threshold. Taking rapid acceleration as an example, setting basic acceleration of each time period of each area of a line, screening out data of which the acceleration is greater than the basic acceleration, calculating an acceleration basic threshold and a duration basic threshold of different areas and time periods by using a box-type diagram outlier method, and judging that the vehicle is rapid acceleration when the acceleration and the duration exceed the acceleration and the duration basic thresholds. The boxplot outlier algorithm is as follows: data are arranged from large to small, an upper quartile Q3, a middle quartile Q1 and a lower quartile Q1 are respectively calculated, two line segments which are the same as the middle line are drawn at Q3+1.5IQR (quartile spacing, IQR is Q3-Q1) and Q1-1.5IQR, the two line segments are outlier cut-off points and are also called upper/lower edges, and the values of the upper edge or the lower edge are used as basic thresholds.
In one embodiment, the apparatus further comprises: and the irregular driving behavior threshold value updating unit 206 is used for recalculating each threshold value of irregular driving behaviors except for large throttle alarm and cold start idling by using the irregular driving behavior threshold value determining unit after the historical data is updated, downloading each updated threshold value to the vehicle-mounted equipment, or judging the region and the time period to which the vehicle-mounted equipment belongs according to real-time line space and time information when the bus line changes, and automatically updating the threshold value by the vehicle-mounted equipment. When the bus route changes, such as the route is temporarily adjusted, or the route A supports the route B, the area and the time period of the bus can be judged according to real-time route space and time information, the threshold value is automatically updated by the vehicle-mounted equipment, the reminding error is prevented, and the safety is improved. This is one of the important points of the present invention.
In one embodiment, the apparatus further comprises: the driver behavior analysis unit 207 counts a value of 80% of data sequence in the state information of the driver in each route space and each time period as a habit value, determines a value exceeding the habit value by a certain proportion as an irregular driving behavior critical value, determines data exceeding the critical value as irregular driving behavior data, and filters misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering device to obtain the filtered irregular driving behavior. The device can filter the misjudgment of the irregular driving behavior so as to improve the accuracy of the evaluation of the driver.
The method comprises the steps of counting a value that 80% of data sequence in state information of drivers in each line space and each time period is a habit value, determining a value exceeding the habit value by a certain proportion to be an irregular driving behavior critical value, and determining data exceeding the critical value to be irregular driving behavior data. Taking rapid acceleration as an example, values of the acceleration and duration of the driver in each region and each time period are respectively calculated, wherein the ranking of the acceleration and duration of each time period is 80%, the acceleration and duration exceeding the values by 20% in the region and the time period are judged to be rapid acceleration critical values, and data exceeding the critical values are irregular driving behavior data. Then, firstly determining a driving scene and then clustering (an HMM classifier can only filter misjudged driver behaviors within a certain intention range), wherein a Hidden Markov Model (HMM) is used for filtering the misjudged driver behaviors: the driving intention corresponds to a hidden state of the hidden markov chain, and the driver behavior is regarded as a composite of the driving operation.
The hidden markov model can be divided into a data acquisition layer, a driving operation classification layer, a driving operation sequence transfer layer and a driver intention transfer layer, wherein the driver intention is a driver stable awareness for a long period of time and belongs to a hidden state in the HMM. The driving operation classification layer is a result of data classification, forms driving operation nodes, describes a certain state and basic behaviors of the current vehicle, and each node represents a certain characteristic of driver operation, so that the driving operation classification layer is convenient to visually understand. And a driving operation sequence transfer layer, wherein the driving operation sequence is a composite result of driving operation and is a comprehensive description of the operation of the driver in a short time to form an observation sequence of the model.
And dividing the observation data of the duration according to time periods and regions, training an HMM classifier by adopting a Baum-Welch algorithm, and randomly assigning values to the hidden state transition probability matrix, the observation state transition probability matrix and the initial state probability matrix.
For example, the rapid acceleration operation is an irregular driving behavior determined in a short time, and therefore it is only possible to determine the driver's intention as a part of the result of the driver's intention execution, and it is not possible to directly determine the driver's intention. And if the sudden deceleration behavior has the consciousness of emergency obstacle avoidance, selecting the vehicle speed, the acceleration and the steering as characteristic vectors, intercepting a part of the continuous observation sequence from a sudden deceleration operation point forwards and backwards, and analyzing the characteristics of the observation sequence. Similar denormal driver behavior clustering: and performing k-means clustering on the irregular driving behaviors in time and space, finding out special reasons in a driving scene, and determining the k value of the irregular driving behavior clustering based on space and time by adopting a device for enumeration and contour coefficient evaluation. When calculating the unnormal score, the deduction weight may be reduced for some special scenes. This is another important inventive point of the present invention.
In one embodiment, the apparatus further comprises: the abnormal driving behavior value calculation unit 208 of the driver calculates the abnormal value of the corresponding driving behavior for each of the filtered irregular driving behaviorsα＝α1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of some index data in the abnormal data, wTIs an abnormal driving behavior critical value.
Setting basic acceleration of each time period of each region of a line by taking rapid acceleration as an example, screening out data of which the acceleration is greater than the basic acceleration, calculating acceleration threshold values and duration threshold values of different regions and time periods by using a box-type diagram outlier method, judging that the vehicle acceleration and the duration exceeding the threshold values are the rapid acceleration, and calculating a global relative deviation value of the rapid acceleration data; respectively calculating values of the driver at 80% of the sequence of the acceleration and the duration of each region and each time period, judging that the acceleration and the duration which exceed the values by 20% in the region and the time period are urgent acceleration critical values, filtering misjudged driver behaviors by an HMM classifier, and then rejecting data of certain scenes by k-means clustering to obtain personal relative deviation values of urgent acceleration; and adding the two relative deviation values to obtain an abnormal value of the driving behavior. This is another important inventive point of the present invention.
In one embodiment, the apparatus further comprises: the bus route and driver evaluation unit 209 counts the irregular behaviors of the bus route space in a period, generates an irregular behavior trend graph of each bus route space, and reminds the lines of which the number of the irregular behaviors exceeds a certain threshold value or the irregular behaviors increase more than a certain percentage on a same ratio; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
The method specifically comprises the following steps: counting the irregular behaviors of each bus route space every day/week, warning the route areas where the irregular behaviors are frequently generated, generating an irregular behavior trend graph of each bus route space, and warning the irregular behaviors to a certain rising degree; and counting all driving behavior abnormal values of each driver on average in each working day, and converting the abnormal values into scores of 60-100 according to the ratio after taking the opposite numbers. And (3) scoring the driving behavior of the driver: and counting the average driving behavior abnormal value (weight problem of different driving behaviors) of the driver in each working day, and converting the abnormal value into a score of 60-100 according to a ratio after taking an inverse number. And ranking the driving behavior of each driver from high to low, and selecting the driver with the score of 90 points or more as an excellent driver, the driver with the score of [70,90) as a good driver and the driver with the score of [60,70) as a driver to be improved.
The device is based on the collected CAN data of the operation of the whole amount of vehicles of all vehicles of a fleet, updates the updated data after dividing the line space and time to automatically generate the irregular driving behavior rule threshold value, updates the irregular driving behavior rule threshold value to the vehicle-mounted equipment in real time, prompts a driver by voice when the irregular driving behavior rule threshold value of the line space and time is exceeded, statistically analyzes the operation trends of the driver and each line, and filters misjudgments in the irregular driving behavior by adopting a Hidden Markov Model (HMM) and a k-means clustering device to prevent the driving behavior of the driver under emergency from being misjudged as the bad driving behavior, so that the traffic safety CAN be improved.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
From the above description of the embodiments, it is clear to those skilled in the art that the present application can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present application may be essentially or partially implemented in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments of the present application.
Finally, it should be noted that: although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that: modifications and equivalents may be made thereto without departing from the spirit and scope of the invention and it is intended to cover in the claims the invention as defined in the appended claims.
Claims (13)
1. A driving safety reminding method based on CAN data is characterized by comprising the following steps:
acquiring historical data, namely acquiring state information and black spot data of a collected bus when the bus runs by connecting vehicle-mounted equipment of a bus through a CAN bus, wherein the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine rotating speed, gear signals, horn signals and door signals, and transmitting the acquired state information to a cloud; the black point data comprises the type, longitude and latitude and black point grade of the black point;
the method comprises the following steps of time and space division, wherein a bus running line and time are divided, the line space is divided according to stations and station intervals, black points are introduced for subdivision, time periods are divided according to working days and weekends every week, and then each day is divided into an early peak, a daytime non-peak section, a late peak and a night non-peak section;
and an irregular driving behavior defining step, wherein the irregular driving behavior is defined as follows: the method comprises the following steps of emergent acceleration, emergent deceleration, emergent stop, emergent acceleration for starting, horn alarm, stable door opening of a vehicle when the vehicle is not stopped, vehicle door opening before starting, neutral sliding, large accelerator alarm, cold start idling, irregular arrival, irregular departure and overspeed, wherein the behavior judgment logics of the large accelerator alarm and the cold start idling are fixed;
the method comprises the following steps of determining an abnormal driving behavior threshold value, namely judging whether black points exist in a line space of the abnormal driving behavior except for large throttle alarm and cold start idling, if the line space is a black point and has hard requirements on the driving behavior, directly using a black point limit value as a threshold value of the abnormal driving behavior, otherwise setting basic threshold values of the abnormal driving behavior in different areas and time periods, wherein the basic threshold values of the abnormal driving behavior in different areas and time periods are data which are screened from historical data and exceed or are lower than the basic values;
and a reminding step, namely when the real-time data in the state information of the running vehicle in the driving process is judged to be the non-standard behavior according to the non-standard driving behavior threshold value, sending a corresponding voice prompt to the driver through the vehicle-mounted equipment.
2. The method of claim 1, wherein the duration of the behavior is also calculated for types of partially non-normative driving behavior, which includes: rapid acceleration, rapid deceleration and overspeed.
3. The method of claim 2, wherein the method comprises:
and an irregular driving behavior threshold value updating step, namely recalculating all threshold values of irregular driving behaviors except for large throttle alarm and cold start idling by using an irregular driving behavior threshold value determining step after the historical data is updated, downloading all updated threshold values to the vehicle-mounted equipment, or judging the region and the time period to which the threshold values belong according to real-time line space and time information when the bus line changes, and automatically updating the threshold values by the vehicle-mounted equipment.
4. The method of claim 3, further comprising:
and analyzing the driver behavior, namely counting that a value of 80% of data sequence of the driver in the state information of each route space and each time period is a habit value, determining that a value exceeding the habit value by a certain proportion is an irregular driving behavior critical value, determining that the data exceeding the critical value is irregular driving behavior data, and filtering misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering method to obtain the filtered irregular driving behavior.
5. The method of claim 4, further comprising:
a driving behavior abnormal value calculation step of calculating an abnormal value α ═ α of the corresponding driving behavior for each of the filtered irregular driving behaviors1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the global abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of certain index data in the personal abnormal data, wTIs an abnormal driving behavior critical value.
6. The method of claim 5, further comprising:
the method comprises the steps of bus route and driver evaluation, wherein the irregular behaviors of bus route spaces in a period are counted, an irregular behavior trend graph of each bus route space is generated, and the lines with the irregular behavior quantity exceeding a certain threshold value or the irregular behaviors rising in proportion exceeding a certain percentage are reminded; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
7. The utility model provides a driving safety reminding device based on CAN data which characterized in that, the device includes:
the system comprises a historical data acquisition unit, a cloud terminal and a cloud terminal, wherein the historical data acquisition unit is connected with vehicle-mounted equipment of a bus through a CAN bus to acquire and collect state information and black spot data of the bus during running, the state information comprises vehicle instrument speed, vehicle acceleration, vehicle track, engine rotating speed, gear signals, horn signals and vehicle door signals, and the acquired state information is transmitted to the cloud terminal; the black point data comprises the type, longitude and latitude and black point grade of the black point;
the time-space division unit is used for dividing the bus running line and time, dividing the line space into lines according to stations and station intervals, introducing black points for subdivision, dividing time periods into time periods according to working days and weekends every week, and dividing each day into early peak, daytime non-peak sections, late peak and night non-peak sections;
the irregular driving behavior defining unit defines irregular driving behaviors as follows: the method comprises the following steps of emergent acceleration, emergent deceleration, emergent stop, emergent acceleration for starting, horn alarm, stable door opening of a vehicle when the vehicle is not stopped, vehicle door opening before starting, neutral sliding, large accelerator alarm, cold start idling, irregular arrival, irregular departure and overspeed, wherein the behavior judgment logics of the large accelerator alarm and the cold start idling are fixed;
the abnormal driving behavior threshold value determining unit is used for judging whether black points exist in the line space of the abnormal driving behavior except for large throttle alarm and cold start idling, if the line space is a black point and has hard requirements on the driving behavior, the black point limit value is directly used as the threshold value of the abnormal driving behavior, otherwise, basic threshold values of the abnormal driving behavior in different areas and time periods are set, and the basic threshold values of the abnormal driving behavior in different areas and time periods are set to be data which are more than or less than the basic values screened from historical data;
and the reminding unit is used for sending corresponding voice prompt to the driver through the vehicle-mounted equipment when the real-time data in the state information of the running vehicle in the driving process is judged to be the non-standard behavior according to the non-standard driving behavior threshold value.
8. The apparatus of claim 7, wherein the duration of the behavior is further calculated for a type of partially non-normative driving behavior, and then the base threshold is calculated using a boxed graph outlier method, the partially non-normative driving behavior comprising: rapid acceleration, rapid deceleration and overspeed.
9. The apparatus of claim 8, wherein the apparatus comprises:
and the irregular driving behavior threshold value updating unit is used for recalculating each threshold value of irregular driving behaviors except for large throttle alarm and cold start idling by using the irregular driving behavior threshold value determining unit after the historical data is updated, downloading each updated threshold value to the vehicle-mounted equipment, or judging the region and the time period to which the threshold value belongs according to real-time line space and time information when the bus line changes, and automatically updating the threshold value by the vehicle-mounted equipment.
10. The apparatus of claim 9, further comprising:
and the driver behavior analysis unit is used for counting that 80% of data sequence of the driver in the state information of each line space and each time period is a habit value, making a value exceeding the habit value by a certain proportion be an irregular driving behavior critical value, making the data exceeding the critical value be irregular driving behavior data, and filtering misjudgment in the irregular driving behavior by using a Hidden Markov Model (HMM) and a k-means clustering device to obtain the filtered irregular driving behavior.
11. The apparatus of claim 10, further comprising:
a driving behavior abnormal value calculation unit for calculating abnormal value of corresponding driving behavior for each filtered abnormal driving behaviorα＝α1+α2；
Wherein,
α1in order to be a global relative deviation value,
wherein liIs the value of some index data in the global abnormal data, lTA threshold value for the corresponding index;
α2in the case of a personal relative deviation value,
wherein wiIs the value of certain index data in the personal abnormal data, wTIs an abnormal driving behavior critical value.
12. The apparatus of claim 11, further comprising:
the bus route and driver evaluation unit is used for counting the irregular behaviors of the bus route space in a period, generating an irregular behavior trend graph of each bus route space and reminding the lines of which the number of the irregular behaviors exceeds a certain threshold value or the irregular behaviors rise more than a certain percentage in a same ratio; and all driving behavior abnormal values of each driver per working day are counted and converted into evaluation scores.
13. A computer-readable storage medium, characterized in that the storage medium has stored thereon computer program code which, when executed by a computer, performs the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910926150.8A CN110615001B (en) | 2019-09-27 | 2019-09-27 | Driving safety reminding method, device and medium based on CAN data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910926150.8A CN110615001B (en) | 2019-09-27 | 2019-09-27 | Driving safety reminding method, device and medium based on CAN data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110615001A CN110615001A (en) | 2019-12-27 |
| CN110615001B true CN110615001B (en) | 2021-04-27 |
Family
ID=68924760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910926150.8A Active CN110615001B (en) | 2019-09-27 | 2019-09-27 | Driving safety reminding method, device and medium based on CAN data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110615001B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111354155B (en) * | 2020-02-25 | 2022-04-12 | 南京领行科技股份有限公司 | Method, device and equipment for carrying out safety reminding in vehicle driving process |
| CN111361568A (en) * | 2020-03-10 | 2020-07-03 | 广州通达汽车电气股份有限公司 | Driver driving behavior evaluation method, device, equipment and storage medium |
| CN111881808B (en) * | 2020-07-23 | 2023-09-15 | 长安大学 | Method for identifying abnormal operation behavior of driver in starting stage of electric bus based on continuous tracking data |
| CN112417945A (en) * | 2020-09-09 | 2021-02-26 | 嘉兴市纳杰微电子技术有限公司 | Distracted driving real-time monitoring method and device based on special neural network |
| CN112328424B (en) * | 2020-12-03 | 2022-05-06 | 之江实验室 | Intelligent anomaly detection method and device for numerical data |
| CN113256064A (en) * | 2021-04-22 | 2021-08-13 | 中国安全生产科学研究院 | Device and method for analyzing driving behavior of subway driver |
| CN113065527B (en) * | 2021-05-07 | 2022-11-18 | 河南科技大学 | Environment interference resistant driving intention online identification method |
| CN114148339B (en) * | 2022-01-17 | 2024-05-17 | 潍柴动力股份有限公司 | Bad driving early warning method and device |
| CN114426025B (en) * | 2022-03-17 | 2023-11-14 | 一汽解放汽车有限公司 | Driving assistance method, driving assistance device, computer device, and storage medium |
| CN115083154A (en) * | 2022-06-13 | 2022-09-20 | 中国第一汽车股份有限公司 | Method and device for detecting vehicle driving event, electronic equipment and medium |
| CN116704780A (en) * | 2023-07-05 | 2023-09-05 | 山东理工大学 | Abnormal driving behavior identification method based on GPS track data |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104417563A (en) * | 2013-09-11 | 2015-03-18 | 丰田自动车株式会社 | Driving assistance device |
| CN204883870U (en) * | 2015-08-19 | 2015-12-16 | 莆田市云驰新能源汽车研究院有限公司 | Driving safety reminds system |
| CN105730444A (en) * | 2016-03-29 | 2016-07-06 | 武汉大学 | Traffic safety control method for highway |
| CN106004883A (en) * | 2016-05-24 | 2016-10-12 | 北京小米移动软件有限公司 | Vehicle violation reminding method and device |
| CN106710144A (en) * | 2016-12-21 | 2017-05-24 | 东软集团股份有限公司 | Driving distance of travel evaluation method and device |
| CN107564292A (en) * | 2017-10-24 | 2018-01-09 | 东南大学 | Passenger's Waiting time rule method of estimation based on website WiFi and vehicle mobile terminals |
| CN108248611A (en) * | 2016-12-29 | 2018-07-06 | 华为技术有限公司 | A kind of method of automatic Pilot, automobile control device, automobile and system |
| CN108346291A (en) * | 2018-04-08 | 2018-07-31 | 谢利明 | Urban Road Traffic Accidents acquisition based on cell phone application and analysis method |
| CN108537198A (en) * | 2018-04-18 | 2018-09-14 | 济南浪潮高新科技投资发展有限公司 | A kind of analysis method of the driving habit based on artificial intelligence |
| CN108860162A (en) * | 2018-07-18 | 2018-11-23 | 平安科技（深圳）有限公司 | Electronic device, safe early warning method and storage medium based on user's driving behavior |
| CN109085837A (en) * | 2018-08-30 | 2018-12-25 | 百度在线网络技术（北京）有限公司 | Control method for vehicle, device, computer equipment and storage medium |
| CN109649395A (en) * | 2018-12-29 | 2019-04-19 | 驭势科技（北京）有限公司 | A kind of automatic Pilot method of adjustment and mobile unit based on computing resource |
| CN109927719A (en) * | 2017-12-15 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | A kind of auxiliary driving method and system based on barrier trajectory predictions |
| CN109979239A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Violation vehicle based reminding method, device and equipment |
| CN110217238A (en) * | 2019-06-18 | 2019-09-10 | 北京中位科技有限公司 | A kind of driving risk class judges optimization method |
-
2019
- 2019-09-27 CN CN201910926150.8A patent/CN110615001B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104417563A (en) * | 2013-09-11 | 2015-03-18 | 丰田自动车株式会社 | Driving assistance device |
| CN204883870U (en) * | 2015-08-19 | 2015-12-16 | 莆田市云驰新能源汽车研究院有限公司 | Driving safety reminds system |
| CN105730444A (en) * | 2016-03-29 | 2016-07-06 | 武汉大学 | Traffic safety control method for highway |
| CN106004883A (en) * | 2016-05-24 | 2016-10-12 | 北京小米移动软件有限公司 | Vehicle violation reminding method and device |
| CN106710144A (en) * | 2016-12-21 | 2017-05-24 | 东软集团股份有限公司 | Driving distance of travel evaluation method and device |
| CN108248611A (en) * | 2016-12-29 | 2018-07-06 | 华为技术有限公司 | A kind of method of automatic Pilot, automobile control device, automobile and system |
| CN107564292A (en) * | 2017-10-24 | 2018-01-09 | 东南大学 | Passenger's Waiting time rule method of estimation based on website WiFi and vehicle mobile terminals |
| CN109927719A (en) * | 2017-12-15 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | A kind of auxiliary driving method and system based on barrier trajectory predictions |
| CN109979239A (en) * | 2017-12-28 | 2019-07-05 | 北京百度网讯科技有限公司 | Violation vehicle based reminding method, device and equipment |
| CN108346291A (en) * | 2018-04-08 | 2018-07-31 | 谢利明 | Urban Road Traffic Accidents acquisition based on cell phone application and analysis method |
| CN108537198A (en) * | 2018-04-18 | 2018-09-14 | 济南浪潮高新科技投资发展有限公司 | A kind of analysis method of the driving habit based on artificial intelligence |
| CN108860162A (en) * | 2018-07-18 | 2018-11-23 | 平安科技（深圳）有限公司 | Electronic device, safe early warning method and storage medium based on user's driving behavior |
| CN109085837A (en) * | 2018-08-30 | 2018-12-25 | 百度在线网络技术（北京）有限公司 | Control method for vehicle, device, computer equipment and storage medium |
| CN109649395A (en) * | 2018-12-29 | 2019-04-19 | 驭势科技（北京）有限公司 | A kind of automatic Pilot method of adjustment and mobile unit based on computing resource |
| CN110217238A (en) * | 2019-06-18 | 2019-09-10 | 北京中位科技有限公司 | A kind of driving risk class judges optimization method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110615001A (en) | 2019-12-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110615001B (en) | Driving safety reminding method, device and medium based on CAN data | |
| US11891071B2 (en) | Vehicle to vehicle telematics | |
| AU2017200782B2 (en) | System and method for in-vehicle operator training | |
| US9747730B2 (en) | Driver measurement and incentive system for improving fuel-efficiency | |
| US9494435B2 (en) | Driving evaluation system and method | |
| EP3239686A1 (en) | Method for determining driving characteristics of a vehicle | |
| CN109263648B (en) | Driving behavior evaluation method, device and equipment | |
| CN113479211B (en) | Method and system for identifying and reminding automobile driving safety behaviors based on machine vision | |
| CN113119985B (en) | Automobile driving data monitoring method, device, equipment and storage medium | |
| US20220051340A1 (en) | System and Method Using Crowd-Sourced Data to Evaluate Driver Performance | |
| CN104732785A (en) | Driving behavior analyzing and reminding method and system | |
| CN111833480B (en) | Driving behavior detection method and device and vehicle | |
| CN104092736A (en) | Vehicle networking device, server and system, scoring method and data collection method | |
| CN112201033B (en) | Road traffic operation active control strategy selection method | |
| CN106022846A (en) | Automobile insurance pricing method, second-hand automobile pricing method and corresponding devices | |
| US9381848B2 (en) | Information provision device for vehicle | |
| CN110723148A (en) | Method and device for identifying bad driving behaviors | |
| CN111209797A (en) | Method, device, equipment and storage medium for monitoring driving behavior | |
| CN112287300A (en) | Data processing method and device, server and storage medium | |
| CN108268678B (en) | Driving behavior analysis method, device and system | |
| CN114084147A (en) | System and method for evaluating driver performance | |
| CN111367968A (en) | Driving data processing method, device, equipment and storage medium | |
| CN114136331B (en) | Driving habit evaluation method, system and detection equipment based on micro-electromechanical gyroscope | |
| Li et al. | Driving type categorizing and microscopic simulation model calibration | |
| CN115966100B (en) | Driving safety control method and system |
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 |