CN105574537A - Multi-sensor-based dangerous driving behavior detection and evaluation method - Google Patents
Multi-sensor-based dangerous driving behavior detection and evaluation method Download PDFInfo
- Publication number
- CN105574537A CN105574537A CN201510813444.1A CN201510813444A CN105574537A CN 105574537 A CN105574537 A CN 105574537A CN 201510813444 A CN201510813444 A CN 201510813444A CN 105574537 A CN105574537 A CN 105574537A
- Authority
- CN
- China
- Prior art keywords
- driving
- vehicle
- driving behavior
- sensor
- waveform
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a multi-sensor-based dangerous driving behavior detection and evaluation method, which comprises the following steps: collecting a plurality of sensor feedback data in a vehicle driving process by a plurality of sensors on a vehicle; analyzing the plurality of sensor feedback data by a default driving behavior recognition algorithm, recognizing a driving mode of the vehicle according to the analysis result and generating a driving behavior analysis result; and comparing the driving behavior analysis result with a default normal driving model, judging whether the driving behavior is dangerous or not and generating a driving behavior evaluation report of the vehicle aiming at each driving behavior and the whole driving course. According to the multi-sensor-based dangerous driving behavior detection and evaluation method, recognition and judgment are carried out on the driving mode of the vehicle which veers, changes a lane and turns tail in the driving process; and the recognition result is compared with a threshold corresponding to a safe driving model by combining with a real-time driving speed, so that the safety of the current driving behavior is judged.
Description
Technical field
The present invention relates to sensor Data Fusion field, particularly a kind of dangerous driving behavioral value based on multisensor and appraisal procedure.
Background technology
Along with the fast lifting of living standard, self-driving automobile has become the main traffic mode of people's trip.Meanwhile, also along with the appearance of dangerous driving behavior.
From driving angle, the threat forming people's trip comprises two aspects: every day all can emerge in large numbers large quantities of driving new hands; Also there is many driving consummations but to the unfamiliar personnel of current friendship rule simultaneously.Therefore driving user needs standard and the guidance that can carry out quantizing to self driving technology judge.In addition, for car insurance industry, more reasonably insurance business can be released targetedly according to user behavior adjustment business demand.By carrying out quantization modulation to the driving procedure of user, the sales mode of vehicle insurance can be rationalized on the one hand, increasing the profit of vehicle insurance business; Also can promote the raising of user's driving ability on the other hand, form good driving attainment.
Further investigation reveals that, except drunk driving and force majeure factor, the cause of most of traffic hazard is the bad steering custom of driving user.Bad driving habits mainly comprise turn to improperly, lane change and turning around.Auxiliary transport condition detection method of driving mainly contains three classes at present:
(1) based on the detection method of OBD (On-BoardDiagnostic): have the highest Stability and veracity, but the OBD interface of different automobile production company and coding difference, it is high that interface reading realizes cost, the security control that simultaneously user may be caused to drive, this makes the technology based on OBD be difficult to popularize;
(2) based on the detection method of video monitoring: this Technical comparing is accurate, but is limited to ambient light and weather effect, the discriminance analysis of great amount of images calculates simultaneously, and calculation cost and carrying cost are very high;
(3) based on the detection method of vehicle-mounted built-in sensors: in some high-end vehicles, expensive sensor, electronic radar and watch-dog is loaded with.These equipment can provide convenient for driving to a certain extent, but generally do not possess the ability to driving behavior analysis.
In addition, for reducing costs and being convenient to popularize, exist in prior art and hope by smart machines such as smart mobile phone, panel computer, vehicle mounted guidance and the vehicle-mounted auxiliary driving of customization, carry out the mode of vehicle detection, but there is defect in various degree respectively.
(1) utilize the post-positioned pick-up head of smart mobile phone to carry out the detection of driving condition according to the information gathering road, but due to hardware environment restriction, real-time and precision are difficult to be satisfied.
(2) utilize the acceleration transducer in smart mobile phone, gyroscope and electronic compass to detect driving behavior, the Space-time Complexity of calculating can be effectively reduced, and be not easy the impact being subject to ambient lighting and weather.But electronic compass is easily subject to the magnetic interference in environment, affects judged result.In addition, acceleration transducer and gyroscope are easily subject to the noise of sensor in the process of data acquisition, are difficult to identify the driving model wherein captured.
Prior art proposes a kind of detecting pattern based on InertialMeasurementUnits (IMUs), can be separated sensor noise to a certain extent, obtains sensor of interest numerical parameter.But the sorted logic of the method to driving behavior is fairly simple, cannot analyze the driving behavior of relative complex.
As shown in Figure 1, prior art also proposes one and utilizes Hall magnetic effect sensor, and namely electronic compass judges turning to of vehicle, measures to turn to behavior in driving.But when having high-intensity magnetic field or electromagnetic interference (EMI) nearby, electronic compass disturbance can be caused even to lose efficacy, and therefore detection method stability is more weak.
Summary of the invention
Object of the present invention is intended at least solve one of described technological deficiency.
For this reason, the object of the invention is to propose a kind of dangerous driving behavioral value based on multisensor and appraisal procedure, align turn under steam, lane change and the driving mode of vehicle that turns around carry out identifying and judging, again by differentiating result, enter to contrast in conjunction with real time running speed threshold value corresponding to safety traffic model, judge the security of current driving behavior.
To achieve these goals, enforcement of the present invention is intended providing a kind of dangerous driving behavioral value based on multisensor and appraisal procedure, comprises the steps:
Step S1, utilizes the multiple sensors on vehicle, gathers the multiple sensors feedback data in described vehicle operation;
Step S2, analyzes with default driving behavior distinguished number described multiple sensors feedback data, identifies the driving mode of described vehicle according to analysis result, generates driving behavior analysis result.Wherein, described driving behavior analysis result comprises the sensor feedback data of each described driving behavior and correspondence;
Step S3, compares described driving behavior analysis result and default normal driving model, judges that whether described driving behavior is dangerous, and for each described driving behavior and the omnidistance report generating the driving behavior assessment of described vehicle of driving.
Further, described multiple sensor comprises: angular-rate sensor and GPS speed pickup.
Further, described multiple sensor adopts one of following two kinds of forms:
(1) angular-rate sensor in described vehicle and GPS speed pickup is fixed on;
(2) be integrated in the gyrostatic angular-rate sensor in smart machine and GPS speed pickup, wherein, described smart machine is fixed on the front window of described vehicle.
Further, multiple sensor feedback data comprise the vehicle speed data of described GPS speed pickup collection and the vehicle of described angular-rate sensor collection turning to, lane change and turn around time waveform sensor data.
Further, in described step S2, described default driving behavior distinguished number adopts MultiWave algorithm.
Further, in described step S2,
First, carry out waved filter to described waveform sensor data, after filtering, Wave data is stored in MultiWave chained list;
Then, Wave data in described MultiWave chained list is analyzed, calculate quantity and the characteristic of correspondence value of efferent echo, according to quantity and the eigenwert identification of described ripple, the driving mode of described vehicle is identified, wherein, described eigenwert comprises: the positive negativity of PN of waveform and, the angular velocity average of maximum angular rate value, single ripple and the duration of single ripple and the absolute value of steering angle.
Further, the waveform sensor data meeting following condition are filtered:
(1) lower than the Wave data of preset angle speed pickup lower threshold;
(2) maximal value of single wave sensor numerical value is lower than the Wave data presetting effective waveform peak threshold value;
(3) waveform duration is lower than the Wave data of default basic waveform duration;
New Wave data not detected yet when exceeding the default detection stand-by period, will be terminated this and detect, and starting to identify driving behavior.
Further, in the described quantity according to ripple and eigenwert identification, the driving mode of described vehicle is identified, comprising::
(1) single ripple: waveform is forward wave, then go out described vehicle and turn left; Waveform is negative wave, then go out described vehicle and turn right; Be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned at interval [155,205] for turning around;
(2) two ripple situations:
When two waveforms are forward wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are left steering; The absolute value of steering angle is positioned at interval [155,205] for turn around left;
When two waveforms are negative wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are right turn; The absolute value of steering angle is positioned at interval [155,205] for turn around to the right;
First waveform is forward wave, and when second waveform is negative wave, described vehicle is lane change left;
First waveform is negative wave, and when second waveform is forward wave, described vehicle is lane change to the right;
(3) three ripples: if be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned near interval [155,205] for turning around;
(4) more than three ripples: the described vehicle short time, violent swing occurred.
Further, in described step S3, preset normal driving model and calculated by the sensor feedback data in the normal vehicle operation under different speed per hour, wherein, described default normal driving model is the driving model by speed per hour classification.
Further, in described step S3, according to the driving behavior judged, the security model of described vehicle speed data with the speed per hour interval of corresponding driving behavior in described default normal driving model is compared, find the security model threshold value in corresponding speed per hour interval, and contrast each data of current form sensor whether in secure threshold interval, if sensor values is within the scope of the security value (s) in corresponding speed per hour interval, then judge described driving behavior safety, otherwise it is dangerous to be judged to described driving behavior.
According to the dangerous driving behavioral value based on multisensor and the appraisal procedure of the embodiment of the present invention, align turn under steam, lane change and the driving mode of vehicle that turns around carry out identifying and judging, and utilize differentiation result, enter to contrast in conjunction with real time running speed threshold value corresponding to safety traffic model, thus judge the security of current driving behavior.
The present invention only needs to call the gyroscope angular-rate sensor in common smart machine and GPS speed pickup, do not need extra hardware device, installation cost is almost nil, and never calls camera, required storage space is low, not by the impact of the link such as weather, light factor.The present invention is to the putting position of equipment also not specific requirement, only need maintenance equipment and vehicle level, in addition the computation complexity of algorithm and power consumption all lower, the CPU of equipment is taken not high, there is good universality, can pass judgment on driving behavior in real time, by log recording, Comprehensive Evaluation be carried out to whole driving behavior simultaneously, possess good traceability.
The aspect that the present invention adds and advantage will part provide in the following description, and part will become obvious from the following description, or be recognized by practice of the present invention.
Accompanying drawing explanation
Above-mentioned and/or additional aspect of the present invention and advantage will become obvious and easy understand from accompanying drawing below to the description of embodiment, wherein:
Fig. 1 is that the ZenDrive of prior art uses the detection of magnetic effect sensor to turn to exemplary plot;
Fig. 2 is according to an embodiment of the invention based on the dangerous driving behavioral value of multisensor and the process flow diagram of appraisal procedure;
Fig. 3 is in accordance with another embodiment of the present invention based on the dangerous driving behavioral value of multisensor and the process flow diagram of appraisal procedure;
Fig. 4 be according to the embodiment of the present invention preset equipment and vehicle relative position schematic diagram;
Fig. 5 is collect left steering waveform example of the dangerous driving behavioral value based on multisensor according to the embodiment of the present invention and appraisal procedure;
Fig. 6 is collect right turn waveform example figure according to the embodiment of the present invention;
Fig. 7 is the once left lane change waveform example figure collected according to the embodiment of the present invention;
Fig. 8 is the once right lane change waveform example figure collected according to the embodiment of the present invention;
Fig. 9 is the waveform example figure that once turns around collected according to the embodiment of the present invention;
Figure 10 be according to the embodiment of the present invention to the threshold value schematic diagram turning to waveform recognition to adopt;
Figure 11 is the threshold value schematic diagram adopted lane change waveform recognition according to the embodiment of the present invention;
Figure 12 is collect complexity many waveforms schematic diagram according to the embodiment of the present invention;
Figure 13 is the schematic diagram of the MultiWave detection finite-state automata according to the embodiment of the present invention;
Figure 14 is according to the direct of the embodiment of the present invention and road and slow also road and comparison of wave shape figure;
Figure 15 is the driving behavior decision flowchart according to the embodiment of the present invention.
Embodiment
Be described below in detail embodiments of the invention, the example of embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has element that is identical or similar functions from start to finish.Be exemplary below by the embodiment be described with reference to the drawings, be intended to for explaining the present invention, and can not limitation of the present invention be interpreted as.
The present invention proposes a kind of dangerous driving behavioral value based on multisensor and appraisal procedure, the method belongs to technology of Internet of things field, in being travelled vehicle in real time by multiple different sensors, the feedback data of each sensor gathers and analyzes, thus judge driver's whether dangerous driving behavior in driving procedure, and assess for its driving procedure.
Fig. 2 is according to an embodiment of the invention based on the dangerous driving behavioral value of multisensor and the process flow diagram of appraisal procedure.
Step S101, obtains the sensor values in vehicle traveling.
Step S102, substitutes into MultiWave detection algorithm and processes by sensor values.
Step S103, judges that the driving behavior of vehicle belongs to left and right turn.
Step S104, judges that the driving behavior of vehicle belongs to left and right lane change.
Step S105, judges that the driving behavior of vehicle belongs to left and right and turns around.
The vehicle speed data that step S106, input GPS speed pickup detects.
Step S107, compares vehicle speed data and default normal driving model.
Step S108, generates and drives result, mark to the driving behavior of vehicle.
As shown in Figure 3, the dangerous driving behavioral value based on multisensor of the embodiment of the present invention and appraisal procedure, comprise the steps:
Step S1, utilizes the multiple sensors on vehicle, the multiple sensors feedback data in collection vehicle driving procedure.
In one embodiment of the invention, multiple sensor comprises: angular-rate sensor and GPS speed pickup.Wherein, multiple sensor adopts one of following two kinds of forms:
(1) angular-rate sensor in vehicle and GPS speed pickup is fixed on;
(2) be integrated in the gyrostatic angular-rate sensor in smart machine and GPS speed pickup, wherein, smart machine is fixed on the front window of vehicle, shown in figure 4.It should be noted that, the position of smart machine is not limited to be arranged on the front window of vehicle, can also be installed on other fixed position places in vehicle, drives visual angle, do not repeat them here as long as meet not hinder.
Multiple sensor feedback data comprise the vehicle speed data of GPS speed pickup collection and the vehicle of angular-rate sensor collection turning to, lane change and turn around time waveform sensor data.Figure 12 is collect complexity many waveforms schematic diagram according to the embodiment of the present invention.
Step S2, multiple sensor feedback data are analyzed with default driving behavior distinguished number, identifies the driving mode of vehicle according to analysis result, generate driving behavior analysis result, wherein, driving behavior analysis result comprises the sensor feedback data of each driving behavior and correspondence.
In one embodiment of the invention, preset driving behavior distinguished number and adopt MultiWave algorithm.Sensing data in the detection vehicle traveling process utilizing MultiWave algorithm real-time, above-mentioned data are filtered and shaping, and according to the Wave data detected, identify driving mode, such as: left/right turns to, left/right lane change, left/right turn around and other behavior.
First carry out sensor preheating, obtain initial offset, for the sensor values of correction tape noise.
Particularly, sensor initial value skew in the equipment caused position is placed for difference, by a preheat mode, pre-read is carried out to sensor values and sensor coordinate system remaps, and statistical study is carried out to initial static number, obtain a sensor values side-play amount caused due to the out-of-flatness of equipment putting position, and revised by the sensor values of side-play amount to later stage Real-time Collection, obtain corresponding waveform sensor data.
Then, carry out waved filter to waveform sensor data, after filtering, Wave data is stored in MultiWave chained list.
By vehicle travel process occur turn to, lane change and turn around time sensor values change, the sensor values that angular-rate sensor obtains is gathered, and minimum sensor threshold value, effectively waveform peak threshold, the effectively parameter such as waveform duration and detection stand-by period by presetting, detect the waveform sensor data formed.By parameters such as the Wave data after filtration and ripple start time, ripple end time, wave duration, maximum sensor numerical value, sensor values average, the positive negativity of ripple in units of ripple, be recorded in the lump in MultiWave chained list.
In one embodiment of the invention, the waveform sensor data meeting following condition are carried out:
(1) lower than the Wave data of preset angle speed pickup lower threshold;
(2) in current single ripple the maximal value of angular-rate sensor lower than presetting the Wave data of effective waveform peak threshold value;
(3) waveform duration is lower than the Wave data of default basic waveform duration;
New Wave data not detected yet when exceeding the default detection stand-by period, will be terminated this and detect, and starting the driving mode identifying vehicle.
In one embodiment of the invention, as shown in figure 14, wave test can be carried out in the following manner: adopt Silence (monitoring), Check (detection), Record (record) and Next (waiting for potential ripple) this one of four states form a finite automaton state machine, in the mode of poll sensor values filtered and process.Table 1 shows the correlation parameter in MultiWave algorithm, and table 2 shows the data structure of single ripple.Figure 10 and Figure 11 respectively illustrates to turn to and describes with lane change waveform threshold value.
Table 1
| Parameter name | Type | Describe |
| wbegin | long | The start time (ms) of ripple |
| wend | long | The end time (ms) of ripple |
| wdur | long | The duration (ms) of ripple |
| wmax | float | The maximal value (rad/s) of this ripple |
| pn | boolean | The positive negativity of ripple |
| speed | float | GPS speed (km/h) during waveform formation |
| wavg | float | Angular-rate sensor average (rad/s) in wave duration |
Table 2
Respectively Silence (monitoring), Check (detection), Record (record) and Next (waiting for potential ripple) one of four states are described below.
(1) Silence (monitoring): continue the numerical value dz monitoring angular-rate sensor, when dz is greater than threshold value Min_dz, jump to Check state; Otherwise continue to monitor.
(2) Check (detection):
2.1), when jumping to Check from Silence: continue the numerical value dz monitoring angular-rate sensor, when sensor values dz is greater than Min_dz, start marking wave, and upgrade when prewave information (table 2); When sensor number dz value is lower than Min_dz, when prewave terminates, judge to work as the condition whether prewave meets a formation ripple by the recorded information of ripple, if met, jump to Record state; Otherwise jump to Silence state to restart.
2.2) time from Next redirect: continue the numerical value dz monitoring angular-rate sensor in Delay time range, if time-out is jumped, forward Next to.If sensor values dz is greater than Min_dz, upgrade when prewave information, and next cycle of poll; When sensor values dz is lower than Min_dz, when prewave terminates, judge to work as the condition whether prewave meets a formation ripple by the recorded information of ripple, jump to Record state if met, otherwise continue poll.
(3) Record (record): the data structure when prewave is detected chained list MultiWave stored in ripple, then enters Next state.
(4) Next (next time): continue to monitor angular-rate sensor numerical value dz in Delay time range, when dz is lower than Min_dz, directly enter next round; Otherwise enter Check state, and start to record a new ripple.If overtime, namely more than Delay time range, then illustrate that an event terminates, and exports MultiWave chained list, jumps to Silence and start to monitor next time.
Then, Wave data in MultiWave chained list is analyzed, calculate quantity and the characteristic of correspondence value of efferent echo, according to quantity and the eigenwert identification of ripple, the driving mode of vehicle is identified, wherein, eigenwert comprises: the angular velocity average of the positive negativity of PN of waveform, maximum angular rate value, single ripple, the duration of single ripple and the absolute value of steering angle.
In one embodiment of the invention, the positive negativity of PN of waveform is calculated by sensor average wavg, wavg>0, then PN is just; Otherwise PN is negative.
Steering angle first by obtaining single ripple steering angle to the sensor number of degrees average of each ripple and wave duration quadrature, then obtains total steering angle to all ripple steering angle summations.
Reference table 2, N is the quantity of ripple, and n is the sequence number of ripple, wavg
nbe the angular-rate sensor average in the n-th wave duration, wdur
nthe duration of to be n-th be ripple.
Particularly, by the positive negativity to the waveform quantity in MultiWave chained list, each ripple, and steering angle carries out comprehensive descision, identifies, comprising the driving mode of vehicle:
Single ripple:
With reference to figure 5 to Fig. 9, turning to, lane change and when turning around, except the difference on direction, complete if this driving behavior is disposable, turn to and turn around only to form a ripple, and wherein left steering and left-hand rotation are turned around formation positive wave, right turn and right-hand rotation are turned around formation negative wave.Lane change can form positive negative two ripples, and being first positive wave during left lane change, is negative wave afterwards; Being first negative wave during right lane change, is positive wave afterwards.
Based on above-mentioned principle, the driving mode of vehicle belonging to single ripple is classified as follows:
Turn left: by the positive negativity PN condition adjudgement travel direction of ripple, PN is timing, when prewave is motion left.That is, waveform is forward wave, then vehicle is turned left.
Turn right: when judging that PN is as time negative, when prewave is motion to the right.That is, waveform is negative wave, then vehicle is turned right.
Turn around: calculate the total angle of rotation θ that this turns to, right-angle steering is 90 degree under normal circumstances, turn around to be 180 degree, but seldom there is perfect driving in general real travel situations, for the fault-tolerance and stability that improve judgement set as follows: if | θ | ∈ [65,115], this turns to and belongs to right-angle steering; If | θ | ∈ [155,205], this turns to belong to and turns around.
The special feelings of a class are had, the driver that some driving habitses are better during lane change, slower in driving Shi Bing road, make the deflection of second ripple not too obvious, specifically see Figure 14, in this case, detection algorithm possibly cannot detect second ripple in lane change traveling in setting threshold value.
Calculate the steering angle of first ripple of left lane change and right lane change, actual | θ | belong to lane change time ∈ [0,20].Therefore, also need to add in single ripple detection logic | θ | ∈ [0,20], judges whether this belongs to lane change situation.
To sum up, be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned at interval [155,205] for turning around.
Situation for two ripples and three ripples:
If to turn to, lane change and disposable completing of turning around to fail, several ripples will be formed, but generally more than 4 ripples, under complex situations, can not can form waveform as shown in figure 12.At this moment waveform differentiates situation more complicated, need ripple comprehensive descision multiple in waveform generation time range, carry out segmentation to the sensor values of vehicle between whole MultiWave recording areas to add up, judged by last total steering angle θ, computing formula is as follows:
For convenience of calculating, directly using the duration that the average of the angular-rate sensor number of degrees is multiplied by ripple to replace formula (2) in the present invention, adopting above-mentioned formula (1) to calculate, repeat no more.
(2) two ripples: two continuous equidirectional ripples may have pause because of in the process turned to or turn around, make angular velocity change have and first increase, then reduce, then increase, finally turned to, again reduced.During lane change, can there is twice contrary slightly turning in direction in vehicle, namely forms the contrary ripple of two positive negativity.For turn left, vehicle first can turn left in motion to, enter left-hand lane, then turn right to, headstock just returns, and sensor values forms the contrary ripple of two crests just.
For these reasons, the driving mode of vehicle belonging to two waveforms is classified as follows:
When two waveforms are forward wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are left steering; The absolute value of steering angle is positioned at interval [155,205] for turn around left;
When two waveforms are negative wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are right turn; The absolute value of steering angle is positioned at interval [155,205] for turn around to the right;
First waveform is forward wave, and when second waveform is negative wave, vehicle is lane change left;
First waveform is negative wave, and when second waveform is forward wave, vehicle is lane change to the right.
Table 3 is the driving mode judged result of two ripples.
Table 3
(3) three ripples: when there being three ripples, directly calculate the angle in whole steering procedure, and are judged the direction that turns to by the positive negativity of steering angle.
For these reasons, the driving mode of vehicle belonging to three waveforms is classified as follows: if be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned at interval [155,205] for turning around.
(4) more than three ripples: when the waveform quantity now in MultiWave chained list is more than or equal to 4, represent that the vehicle short time, violent swing occurred, driver does not have steering vehicle stably to keep straight on or rationally turn to, and the vehicle body of swing forms certain threat to nearby vehicle, belongs to dangerous driving.
Figure 15 is the driving behavior decision flowchart according to the embodiment of the present invention.
Step S1501, reads the data in MultiWave chained list.
Step S1502, judges the quantity of ripple.
Step S1503, judge the quantity of ripple be 1 or 2 time, judge the positive and negative shape PN of waveform.
Step S1504, when 1 or 2 ripple, according to judging that vehicle is turned left or turned right, and when the quantity judging ripple is 3, judges the absolute value of steering angle further.
Step S1505, if the absolute value of steering angle is positioned at [0,20], then judges that the driving behavior of vehicle belongs to lane change;
Step S1506, if the absolute value of steering angle is positioned at [65,115], then judges that the driving behavior of vehicle belongs to and turns to;
Step S1507, if the absolute value of steering angle is positioned at [155,205], then judges that the driving behavior of vehicle belongs to and turns around;
Step S1508, when judging the quantity of ripple more than 3, judges that the driving behavior of vehicle belongs to dangerous driving.
Step S3, driving behavior analysis result and default normal driving model are compared, judge that whether driving behavior is dangerous, and for each driving behavior with drive the omnidistance driving behavior assessment report generating vehicle, utilize the various features of waveform to carry out to vehicle running state identifications and be that this driving procedure is marked according to code of points.
First preset normal driving model to be below described.
In one embodiment of the invention, preset normal driving model to be calculated by the sensor feedback data in the normal vehicle operation under different speed per hour.Wherein, default normal driving model is the driving model by speed per hour classification.
It should be noted that, current normal driving model to be driven at low speed situation modeling mainly for urban district, and normal driving model constantly can be added by the later stage, makes model more perfect, enables to process more travel situations.
Particularly, by being fixed on the smart machine of vehicle front windshield, to record in different preset vehicle speed situation that vehicle carries out turning at conventional city road, lane change and turn around time, the data such as the positive negativity of the extreme value (wmax) of the waveform quantity that the angular-rate sensor numerical value in gyroscope is formed, waveform duration (wdur), wave numerics, angular-rate sensor average (wavg) and ripple.Wherein wdur unit is millisecond (ms), wmax and wavg unit is degree (rad/s) per second.Then by a large amount of data acquisitions and contrast, determine to turn to, lane change and the security model threshold value that turns around, generate corresponding security parameter interval, as shown in table 4.
Table 4
Find after tested, angular-rate sensor numerical value under normal road condition during right-angle steering is worth and is generally greater than 0.25rad/s most, interval [0.29,0.76] (rad/s), wave duration is [3600,7500] (ms), and turning velocity generally cannot be greater than 30km/h, namely speed per hour turns to behavior more dangerous more than 30km/h.
Meanwhile, speed per hour and wave duration are inversely proportional to, and be worth most be directly proportional with average to angular velocity.Angular velocity is to linear velocity conversion formula (3), and wherein v is linear velocity, and w is angular velocity, and r is turning radius.
From formula (3), actual time line speed is directly proportional to angular velocity in theory, consistent with test result.Simultaneously, under the condition that speed per hour is identical, angular-rate sensor numerical value during right turn is worth and is generally greater than left-hand rotation most, wave duration is lower than the left-hand rotation wave duration simultaneously, this also meets the situation that Ordinary Rd crossing left steering radius is greater than right turn radius, concrete numerical value is in table 4, Fig. 5 and Fig. 6 is respectively the oscillogram that left and right turn is formed.
v＝wr(3)
With reference to figure 4 and Fig. 5, for lane change, to be generally under normal circumstances made up of two continuous, contrary ripples: left lane change is made up of a forward wave and a negative wave; Right lane change is made up of a negative wave and a forward wave.It is not obvious that lane change waveform affects change by real-time speed, but maximum angular rate numerical value and wave duration are all little than turning to numerical value, and concrete numerical value is in table 5.In addition, the numerical value average of forward wave and negative wave and the product of wave duration and be generally approximately zero, calculate as formula (4), wherein wavg is for turning to waveform average, and wdur is for turning to waveform duration.
wavg
+*wdur
++wavg
-*wdur
-≈0(4)
Table 5
For turning around, except in the process of turning around, angular velocity changes greatly, the average of angular-rate sensor, and the T.T. that waveform continues is equally all very large.Under angular velocity average when turning around and high-speed case, turning rate is more close, but the wave duration is the longest, and at about 10s, total steering angle θ is about 180 degree, reference table 6.
| Wmax | Wavg | Wdur | θ 10 --> | |
| Turn around in a left side | 0.5 | 0.31 | 9510 | 168 |
| Turn around in the right side | -0.56 | -0.33 | 10063 | -177 |
Table 6
According to analyze in step S2 obtain turn to, lane change and the result that turns around, the vehicle speed data that the waveform obtained in conjunction with GPS speed pickup starts, be mapped to the driving behavior category of model in corresponding speed interval, Data Comparison is carried out to when front steering data and default normal driving behavior model, judge whether in the reasonable scope each value, if every data of current driving waveform are in security value (s) interval range, just think safety, otherwise think dangerous.
Particularly, according to the driving behavior judged, the security model of vehicle speed data with the speed per hour interval of corresponding driving behavior in default normal driving model is compared, find the security model threshold interval in corresponding speed per hour interval, and contrast each data of current form sensor whether in secure threshold interval, then judge driving behavior safety, otherwise judge that driving behavior is dangerous.
First real-time safe scoring is carried out to single driving event, then omnidistance overall behavior is driven to certain and carry out comprehensive grading.
(1) single driving behavior scoring, the Quantitative marking to each driving behavior:
In single driving behavior scoring, the secure threshold under friction speed fluctuates 20% as security interval, to round up process, then be mapped to security interval to a real-time speed position.
Code of points: use 5 points to make scoring and each driving behavior is passed judgment on.Main factor of judgment comprises: the average (wavg) of ripple, the most value (wmax) of ripple and wave duration (wdur), code of points is in table 7.Wherein in single ripple situation turn to and turn around directly to use waveform average, waveform be worth most and waveform duration calculate; When being greater than a ripple, waveform duration adds up as last total duration, and waveform average and waveform are worth the mean value computation using multiple ripple most.For lane change, average to after the calculating respectively of former and later two ripples.The note that single parameter belongs to corresponding security parameter interval is once effective.0 point is directly judged for the situation of wave number amount more than 3.
| Effective degree | Score value |
| 3 | 5 |
| 2 | 4 |
| 1 | 2 |
| 0 | 1 |
| Wave number amount >=4 | 0 |
Table 7
(2) this drives comprehensive grading, carries out comprehensive grading to a series of driving behavior in a driving procedure.
Record is carried out in single driving behavior and score value, and finally total score value gets each average of driving scoring, and more than 3 points is qualified, defective lower than 3 points.Calculate the occurrence number of each driving behavior, and the qualified ratio of each behavior.And to score value lower than 3 points, and behavior accounts for the driving behavior that total driving behavior quantity is greater than 20% carries out recording and exporting, and needs the driving behavior carrying out paying close attention to and improving for pointing out driver.
To sum up, the present invention can utilize multinomial sensor to catch the turning to of vehicle, lane change and the event such as to turn around, and analyzes potential dangerous driving behavior, and provides a set of driving scoring.
The present invention adopts many moneys high-, middle-and low-end smart machine to detect vehicle, detected type of vehicle comprises pony car three and SUV two, participate in tester and comprise special train driver, taxi driver and private car car owner, drive about 40 hours altogether test duration.Test SUV vehicle driving event 336 times; Special train car 462 times; Taxi car 420 times.
Test result is that left and right turn detects bat 100%, and lane change detects bat 89%, turns around to detect bat 93%.
According to test result analysis: low side smart machine discrimination is lower, is about 85%, turn around discrimination about 90%; Middle and high end smart machine (Samsung s3, Samsung s4, association S850 and Nexus6) discrimination is all about 90%, and the discrimination that turns around is about 94%.Compared with the type of middle and high end, the detection perform of low side smart mobile phone affects comparatively large by sensor and CPU computing power, every discrimination is lower.
According to the dangerous driving behavioral value based on multisensor and the appraisal procedure of the embodiment of the present invention, align turn under steam, lane change and the driving mode of vehicle that turns around carry out identifying and judging, and utilize differentiation result, enter to contrast in conjunction with real time running speed threshold value corresponding to safety traffic model, thus judge the security of current driving behavior.
The present invention only needs to call the gyroscope angular-rate sensor in common smart machine and GPS speed pickup, do not need extra hardware device, installation cost is almost nil, and never calls camera, required storage space is low, not by the impact of the link such as weather, light factor.The present invention is to the putting position of equipment also not specific requirement, only need maintenance equipment and vehicle level, in addition the computation complexity of algorithm and power consumption all lower, the CPU of equipment is taken not high, there is good universality, can pass judgment on driving behavior in real time, by log recording, Comprehensive Evaluation be carried out to whole driving behavior simultaneously, possess good traceability.
In the description of this instructions, specific features, structure, material or feature that the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to describe in conjunction with this embodiment or example are contained at least one embodiment of the present invention or example.In this manual, identical embodiment or example are not necessarily referred to the schematic representation of above-mentioned term.And the specific features of description, structure, material or feature can combine in an appropriate manner in any one or more embodiment or example.
Although illustrate and describe embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, those of ordinary skill in the art can change above-described embodiment within the scope of the invention when not departing from principle of the present invention and aim, revising, replacing and modification.Scope of the present invention is by claims extremely equivalency.
Claims (10)
1., based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, comprise the steps:
Step S1, utilizes the multiple sensors on vehicle, gathers the multiple sensors feedback data in described vehicle operation;
Step S2, described multiple sensors feedback data is analyzed with default driving behavior distinguished number, the driving mode of described vehicle is identified according to analysis result, generate driving behavior analysis result, wherein, described driving behavior analysis result comprises the sensor feedback data of each described driving behavior and correspondence;
Step S3, compares described driving behavior analysis result and default normal driving model, judges that whether described driving behavior is dangerous, and for each described driving behavior and the driving behavior assessment report driving the described vehicle of omnidistance generation.
2., as claimed in claim 1 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, described multiple sensor comprises: angular-rate sensor and GPS speed pickup.
3. as claimed in claim 2 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, described multiple sensor adopts one of following two kinds of forms:
(1) angular-rate sensor in described vehicle and GPS speed pickup is fixed on;
(2) be integrated in the gyrostatic angular-rate sensor in smart machine and GPS speed pickup, wherein, described smart machine is fixed on described vehicle front window.
4. as claimed in claim 3 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, multiple sensor feedback data comprise the vehicle speed data of described GPS speed pickup collection and the vehicle of described angular-rate sensor collection turning to, lane change and turn around time waveform sensor data.
5. as claimed in claim 4 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, in described step S2, described default driving behavior distinguished number adopts MultiWave algorithm.
6., as claimed in claim 5 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, in described step S2,
First, carry out waved filter to described waveform sensor data, after filtering, Wave data is stored in MultiWave chained list;
Then, Wave data in described MultiWave chained list is analyzed, calculate quantity and the characteristic of correspondence value of efferent echo, according to quantity and the eigenwert identification of described ripple, the driving mode of described vehicle is identified, wherein, described eigenwert comprises: the angular velocity average of the positive negativity of PN of waveform, maximum angular rate value, single ripple, the duration of single ripple and the absolute value of steering angle.
7., as claimed in claim 6 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, filter out the waveform sensor data meeting following condition:
(1) lower than the Wave data of preset angle speed pickup lower threshold;
(2) maximal value of single wave sensor numerical value is lower than the Wave data presetting effective waveform peak threshold value;
(3) waveform duration is lower than the Wave data of default basic waveform duration;
New Wave data not detected yet when exceeding the default detection stand-by period, will be terminated this and detect, and starting to identify driving behavior.
8. as claimed in claim 6 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, in the described quantity according to ripple and eigenwert identification, the driving mode of described vehicle identified, comprising::
(1) single ripple: waveform is forward wave, then go out described vehicle and turn left; Waveform is negative wave, then described vehicle is turned right; Be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned at interval [155,205] for turning around;
(2) two ripples:
When two waveforms are forward wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are left steering; The absolute value of steering angle is positioned at interval [155,205] for turn around left;
When two waveforms are negative wave, continue to judge steering angle, if the absolute value of steering angle is positioned at interval, [65,115] are right turn; The absolute value of steering angle is positioned at interval [155,205] for turn around to the right;
First waveform is forward wave, and when second waveform is negative wave, described vehicle is lane change left;
First waveform is negative wave, and when second waveform is forward wave, described vehicle is lane change to the right;
(3) three ripples: if be gentle lane change when the absolute value of steering angle is positioned at interval [0,20]; The absolute value of steering angle is positioned at interval [65,115] for turning to; The absolute value of steering angle is positioned at interval [155,205] for turning around;
(4) more than three ripples: the described vehicle short time, violent swing occurred.
9. as claimed in claim 4 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, in described step S3, preset normal driving model to be calculated by the sensor feedback data in the normal vehicle operation under different speed per hour, wherein, described default normal driving model is the driving model by speed per hour classification.
10. as claimed in claim 9 based on dangerous driving behavioral value and the appraisal procedure of multisensor, it is characterized in that, in described step S3, according to the driving behavior judged, the security model of described vehicle speed data with the speed per hour interval of corresponding driving behavior in described default normal driving model is compared, find the security model threshold value in corresponding speed per hour interval, and contrast each data of current form sensor whether in secure threshold interval, if sensor values is within the scope of the security value (s) in corresponding speed per hour interval, then judge described driving behavior safety, otherwise judge that described driving behavior is dangerous.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510813444.1A CN105574537B (en) | 2015-11-23 | 2015-11-23 | Dangerous driving behavior testing and evaluation method based on multisensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510813444.1A CN105574537B (en) | 2015-11-23 | 2015-11-23 | Dangerous driving behavior testing and evaluation method based on multisensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105574537A true CN105574537A (en) | 2016-05-11 |
| CN105574537B CN105574537B (en) | 2018-12-28 |
Family
ID=55884643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510813444.1A Expired - Fee Related CN105574537B (en) | 2015-11-23 | 2015-11-23 | Dangerous driving behavior testing and evaluation method based on multisensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105574537B (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106627871A (en) * | 2016-12-15 | 2017-05-10 | 重庆典果成实业有限公司 | Safety auxiliary system for motorcycle |
| CN106682295A (en) * | 2016-12-16 | 2017-05-17 | 清华大学 | Analysis method for evaluating real-time safety characteristics of drivers |
| CN106740852A (en) * | 2016-12-15 | 2017-05-31 | 重庆典果成实业有限公司 | The method that safe auxiliary is carried out during for motorcycle turns |
| CN106882141A (en) * | 2017-02-16 | 2017-06-23 | 深圳市航天无线通信技术有限公司 | Driving behavior detection method and device |
| CN107024358A (en) * | 2017-05-23 | 2017-08-08 | 徐工集团工程机械股份有限公司 | A kind of operating loading analysis system based on multiple-sensor network |
| CN107444404A (en) * | 2017-07-25 | 2017-12-08 | 武汉依迅北斗空间技术有限公司 | A kind of evaluation method and device of driving behavior standard degree |
| CN108122298A (en) * | 2016-11-30 | 2018-06-05 | 厦门雅迅网络股份有限公司 | Improve the method and system of steering wheel for vehicle angle data accuracy |
| CN109002797A (en) * | 2018-07-16 | 2018-12-14 | 腾讯科技（深圳）有限公司 | Vehicle lane change detection method, device, storage medium and computer equipment |
| CN109754595A (en) * | 2017-11-01 | 2019-05-14 | 阿里巴巴集团控股有限公司 | Appraisal procedure, device and the interface equipment of vehicle risk |
| CN109878530A (en) * | 2019-02-28 | 2019-06-14 | 中国第一汽车股份有限公司 | Identify the method and system of the lateral driving cycle of vehicle |
| CN110211401A (en) * | 2018-02-28 | 2019-09-06 | 长城汽车股份有限公司 | A kind of dangerous tip method and device of driving |
| CN110692094A (en) * | 2017-06-02 | 2020-01-14 | 本田技研工业株式会社 | Vehicle control apparatus and method for control of autonomous vehicle |
| CN110832569A (en) * | 2017-07-05 | 2020-02-21 | 松下知识产权经营株式会社 | System and method for detecting cheating of an autonomous vehicle in transit |
| WO2020169052A1 (en) * | 2019-02-21 | 2020-08-27 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for driving condition identification |
| CN111783536A (en) * | 2020-05-29 | 2020-10-16 | 北京百度网讯科技有限公司 | Method, device and equipment for evaluating lane change reasonability and storage medium |
| CN111986489A (en) * | 2019-11-07 | 2020-11-24 | 深圳鼎然信息科技有限公司 | Intelligent safety early warning method, device and system for vehicle and storage medium |
| CN112287797A (en) * | 2020-10-23 | 2021-01-29 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
| CN114172705A (en) * | 2021-11-29 | 2022-03-11 | 北京智美互联科技有限公司 | Network big data analysis method and system based on pattern recognition |
| CN114221794A (en) * | 2021-11-29 | 2022-03-22 | 北京智美互联科技有限公司 | Feature analysis method and system using improved pattern recognition algorithm |
| US11300961B2 (en) | 2017-06-02 | 2022-04-12 | Honda Motor Co., Ltd. | Vehicle control apparatus and method for controlling automated driving vehicle |
| CN115063055A (en) * | 2022-08-17 | 2022-09-16 | 山东宇翔车桥制造有限公司 | Semitrailer equipment operation performance supervisory systems based on data analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463244A (en) * | 2014-12-04 | 2015-03-25 | 上海交通大学 | Aberrant driving behavior monitoring and recognizing method and system based on smart mobile terminal |
| CN104527647A (en) * | 2014-12-15 | 2015-04-22 | 清华大学 | Monitoring and evaluation method of driving behavior risk degrees |
| CN104802737A (en) * | 2015-03-25 | 2015-07-29 | 清华大学 | Mobile phone based vehicle abnormality driving behavior detection method |
-
2015
- 2015-11-23 CN CN201510813444.1A patent/CN105574537B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463244A (en) * | 2014-12-04 | 2015-03-25 | 上海交通大学 | Aberrant driving behavior monitoring and recognizing method and system based on smart mobile terminal |
| CN104527647A (en) * | 2014-12-15 | 2015-04-22 | 清华大学 | Monitoring and evaluation method of driving behavior risk degrees |
| CN104802737A (en) * | 2015-03-25 | 2015-07-29 | 清华大学 | Mobile phone based vehicle abnormality driving behavior detection method |
Non-Patent Citations (1)
| Title |
|---|
| 詹彤 等: "基于多传感器信息融合的危险行为检测系统", 《汽车零部件》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108122298A (en) * | 2016-11-30 | 2018-06-05 | 厦门雅迅网络股份有限公司 | Improve the method and system of steering wheel for vehicle angle data accuracy |
| CN108122298B (en) * | 2016-11-30 | 2021-06-29 | 厦门雅迅网络股份有限公司 | Method and system for improving accuracy of steering wheel angle data of vehicle |
| CN106740852A (en) * | 2016-12-15 | 2017-05-31 | 重庆典果成实业有限公司 | The method that safe auxiliary is carried out during for motorcycle turns |
| CN106627871A (en) * | 2016-12-15 | 2017-05-10 | 重庆典果成实业有限公司 | Safety auxiliary system for motorcycle |
| CN106682295A (en) * | 2016-12-16 | 2017-05-17 | 清华大学 | Analysis method for evaluating real-time safety characteristics of drivers |
| CN106882141A (en) * | 2017-02-16 | 2017-06-23 | 深圳市航天无线通信技术有限公司 | Driving behavior detection method and device |
| CN106882141B (en) * | 2017-02-16 | 2020-02-21 | 深圳市航天无线通信技术有限公司 | Driving behavior detection method and device |
| CN107024358B (en) * | 2017-05-23 | 2019-03-01 | 徐工集团工程机械股份有限公司 | A kind of operating loading analysis system based on multiple-sensor network |
| CN107024358A (en) * | 2017-05-23 | 2017-08-08 | 徐工集团工程机械股份有限公司 | A kind of operating loading analysis system based on multiple-sensor network |
| US11300961B2 (en) | 2017-06-02 | 2022-04-12 | Honda Motor Co., Ltd. | Vehicle control apparatus and method for controlling automated driving vehicle |
| US11275379B2 (en) | 2017-06-02 | 2022-03-15 | Honda Motor Co., Ltd. | Vehicle control apparatus and method for controlling automated driving vehicle |
| CN110692094A (en) * | 2017-06-02 | 2020-01-14 | 本田技研工业株式会社 | Vehicle control apparatus and method for control of autonomous vehicle |
| CN110832569A (en) * | 2017-07-05 | 2020-02-21 | 松下知识产权经营株式会社 | System and method for detecting cheating of an autonomous vehicle in transit |
| CN107444404A (en) * | 2017-07-25 | 2017-12-08 | 武汉依迅北斗空间技术有限公司 | A kind of evaluation method and device of driving behavior standard degree |
| CN109754595B (en) * | 2017-11-01 | 2022-02-01 | 阿里巴巴集团控股有限公司 | Vehicle risk assessment method and device and interface equipment |
| CN109754595A (en) * | 2017-11-01 | 2019-05-14 | 阿里巴巴集团控股有限公司 | Appraisal procedure, device and the interface equipment of vehicle risk |
| CN110211401A (en) * | 2018-02-28 | 2019-09-06 | 长城汽车股份有限公司 | A kind of dangerous tip method and device of driving |
| CN109002797A (en) * | 2018-07-16 | 2018-12-14 | 腾讯科技（深圳）有限公司 | Vehicle lane change detection method, device, storage medium and computer equipment |
| CN109002797B (en) * | 2018-07-16 | 2023-07-25 | 腾讯科技（深圳）有限公司 | Vehicle lane change detection method, device, storage medium and computer equipment |
| WO2020169052A1 (en) * | 2019-02-21 | 2020-08-27 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for driving condition identification |
| CN109878530A (en) * | 2019-02-28 | 2019-06-14 | 中国第一汽车股份有限公司 | Identify the method and system of the lateral driving cycle of vehicle |
| CN111986489A (en) * | 2019-11-07 | 2020-11-24 | 深圳鼎然信息科技有限公司 | Intelligent safety early warning method, device and system for vehicle and storage medium |
| CN111986489B (en) * | 2019-11-07 | 2021-08-03 | 深圳鼎然信息科技有限公司 | Intelligent safety early warning method, device and system for vehicle and storage medium |
| CN111783536A (en) * | 2020-05-29 | 2020-10-16 | 北京百度网讯科技有限公司 | Method, device and equipment for evaluating lane change reasonability and storage medium |
| CN112287797A (en) * | 2020-10-23 | 2021-01-29 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
| CN114221794A (en) * | 2021-11-29 | 2022-03-22 | 北京智美互联科技有限公司 | Feature analysis method and system using improved pattern recognition algorithm |
| CN114172705A (en) * | 2021-11-29 | 2022-03-11 | 北京智美互联科技有限公司 | Network big data analysis method and system based on pattern recognition |
| CN114172705B (en) * | 2021-11-29 | 2023-11-14 | 北京国瑞数智技术有限公司 | Network big data analysis method and system based on pattern recognition |
| CN114221794B (en) * | 2021-11-29 | 2024-02-02 | 北京国瑞数智技术有限公司 | User behavior analysis method, system and medium based on improved pattern recognition algorithm |
| CN115063055A (en) * | 2022-08-17 | 2022-09-16 | 山东宇翔车桥制造有限公司 | Semitrailer equipment operation performance supervisory systems based on data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105574537B (en) | 2018-12-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105574537A (en) | Multi-sensor-based dangerous driving behavior detection and evaluation method | |
| US12106661B2 (en) | Determining causation of traffic events and encouraging good driving behavior | |
| CN103770733B (en) | Method and device for detecting safety driving states of driver | |
| US10304260B2 (en) | Automated vehicle identification and inspection data processing system | |
| CN105761149A (en) | Vehicle insurance system and vehicle insurance charging method | |
| CN102717765B (en) | Fatigue driving detection method and anti-fatigue driving auxiliary device | |
| CN103043057B (en) | Abnormal driving based on vehicle position information judges and warning system | |
| CN103767715B (en) | A kind of driver safety driving state detection device | |
| CN105564436A (en) | Advanced driver assistance system | |
| US20160110650A1 (en) | Advanced context-based driver scoring | |
| Castignani et al. | An evaluation study of driver profiling fuzzy algorithms using smartphones | |
| US12065138B2 (en) | Vehicle surface impact detection | |
| CN109389864A (en) | A kind of vehicle relative positioning and anti-collision warning method | |
| CN105599773A (en) | Driver state indicating device and method based on vehicle moving state | |
| CN103093647A (en) | Method for operating a vehicle system of a motor vehicle and motor vehicle | |
| CN104794855A (en) | Driver's attention comprehensive assessment system | |
| CN104021370A (en) | Driver state monitoring method based on vision information fusion and driver state monitoring system based on vision information fusion | |
| Sun et al. | An integrated solution for lane level irregular driving detection on highways | |
| CN104794856A (en) | Driver's attention comprehensive evaluation system | |
| CN106209546A (en) | Based on binocular camera and area array cameras automatic with car system | |
| CN106373461A (en) | Driver training management service system and method | |
| CN103489010A (en) | Fatigue driving detecting method based on driving behaviors | |
| Wu et al. | L3: Sensing driving conditions for vehicle lane-level localization on highways | |
| CN112967497B (en) | Shared automobile supervision system and supervision method | |
| CN108482385B (en) | A kind of mobile unit operational risk discrimination method considering traffic environment complexity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181228 Termination date: 20201123 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |