CN111506692B - Collision detection method based on target behaviors - Google Patents
Collision detection method based on target behaviors Download PDFInfo
- Publication number
- CN111506692B CN111506692B CN202010316688.XA CN202010316688A CN111506692B CN 111506692 B CN111506692 B CN 111506692B CN 202010316688 A CN202010316688 A CN 202010316688A CN 111506692 B CN111506692 B CN 111506692B
- Authority
- CN
- China
- Prior art keywords
- vector
- target behavior
- triaxial
- collision detection
- order difference
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Remote Sensing (AREA)
- Alarm Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a collision detection method based on target behaviors, which comprises the following steps: constructing a current target behavior entity description, and acquiring a historical target behavior entity description set; when a historical target behavior entity description set exists, similarity between the current target behavior entity description and each historical target behavior entity description in the historical target behavior entity description set is calculated respectively, a plurality of historical target behavior entity descriptions with the front similarity are obtained, statistics of the plurality of historical target behavior entity descriptions are extracted, and collision detection is carried out based on the current target behavior entity description and the statistics; and when the historical target behavior entity description set does not exist, performing collision detection based on the current target behavior entity description. For the user or the user population, the invention utilizes the historical behavior habit clusters to carry out collision detection, can avoid false alarm or missing report, and greatly improves the collision detection accuracy with personalized driving habits.
Description
Technical Field
The invention relates to the field of Internet of vehicles, in particular to a collision detection method based on historical behaviors of targets (user groups or single users).
Background
In the collision detection technology based on track data, the following drawbacks exist when the collision detection is performed by methods such as threshold value setting, statistical analysis, machine learning, historical behavior analysis and the like:
1. the method relies on a large amount of marking data, and the total assumption of the track data is consistent and distributed, so that when the new installation of the equipment is not considered, the track data is not relied on, and the false alarm and the missing report of the collision are caused.
2. For car owners of the same group, the behaviors of the group are not researched at present, so that after the group migration occurs, a model is not followed, and false alarm and missing report of collision are caused.
3. For a single user, track data with abnormal statistics values and higher than a threshold value is quite possibly caused by personalized differences of the vehicle and the vehicle owners, and the existing method aims at analyzing all users at the same time without difference, so that a lot of false positives are generated.
Therefore, it is desirable to devise a completely new approach to collision detection based on characteristics of the user or the population itself.
Disclosure of Invention
The invention aims at: aiming at the problems, the collision detection method based on the target behaviors is provided to more accurately judge whether the current vehicle collides or not based on the behaviors of the user or the user population and the personalized characteristics of the behaviors.
The technical scheme adopted by the invention is as follows:
a collision detection method based on target behavior, comprising the following processes:
constructing a current target behavior entity description, and acquiring a historical target behavior entity description set;
when a historical target behavior entity description set exists, similarity between the current target behavior entity description and each historical target behavior entity description in the historical target behavior entity description set is calculated respectively, a plurality of historical target behavior entity descriptions with the front similarity are obtained, statistics of the plurality of historical target behavior entity descriptions are extracted, and collision detection is carried out based on the current target behavior entity description and the statistics;
when the historical target behavior entity description set does not exist, collision detection is carried out based on the current target behavior entity description;
the construction method of the current target behavioral entity description comprises the following A, B, C processes:
A. analyzing vehicle motion parameters of preset dimensions in the track data packet of preset duration/length from the track data packet uploaded by the equipment end in real time, and vectorizing the vehicle motion parameters of each dimension;
B. carrying out first preprocessing on the vectorized analysis result to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension;
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to construct the target behavior entity description.
Further, the construction method of the historical target behavior entity description set comprises the following steps:
A. extracting a track data packet to be clustered of a target;
and respectively executing B-D for each track data packet:
B. analyzing vehicle motion parameters of preset dimensions from each point of the track data packet, and carrying out vectorization processing on the vehicle motion parameters of each dimension;
C. carrying out first preprocessing on the vectors of each dimension to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension;
D. performing second preprocessing on each dimension parameter in the feature set to construct a target behavior entity description;
E. constructing a similarity matrix according to the similarity between the descriptions of each target behavior entity;
F. clustering the target behaviors according to the similarity matrix;
G. and calculating the central point and the related statistic of each target behavior cluster, and storing the central point and the related statistic of each target behavior cluster in a correlated way.
In summary, due to the adoption of the technical scheme, the beneficial effects of the invention are as follows:
1. according to the method and the device for identifying the collision of the user population, collision occurrence probability of the user population during population migration is calculated, and personalized driving behaviors, such as high acceleration, of the user caused by population migration are prevented from being identified as collision, so that accuracy is improved. For a single user, according to the clustering result of the historical driving behaviors, false alarm caused by no personalized reference during the collision detection of the historical behavior clusters constructed according to the overall data (big data) can be avoided.
2. According to the invention, collision detection (collision probability calculation) is carried out on the user or the population according to the historical driving habit of the user or the population, so that the missing report that the user data is normal and collision actually occurs during detection according to the overall data (big data) is avoided, and the recall rate of the collision detection can be improved.
Drawings
The invention will now be described by way of example and with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of a method of collision detection based on target behavior.
FIG. 2 is a flow chart of a historical target behavioral entity description construct.
Detailed Description
All of the features disclosed in this specification, or all of the steps in a method or process disclosed, may be combined in any combination, except for mutually exclusive features and/or steps.
Any feature disclosed in this specification (including any accompanying claims, abstract) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. That is, each feature is one example only of a generic series of equivalent or similar features, unless expressly stated otherwise.
Example 1
The embodiment discloses a target behavior entity description construction method, which comprises the following steps:
A. And analyzing the data packets of each point in the track section with the preset duration/length from the track data packets uploaded by the equipment end in real time, and respectively carrying out vectorization processing on the analysis results with specific dimensions.
B. Carrying out first preprocessing on the vectorized analysis result to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension.
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to construct the target behavior entity description.
The track data packets analyzed are different according to the different target objects. For the user population targets, the behaviors of a plurality of users at the same time are required to be analyzed, and the acquired track data packet is the track data packet which is acquired and uploaded by a plurality of equipment ends at the same time, namely multi-user transverse analysis. For a single user, the current behavior of the user needs to be analyzed, and the acquired track data packet is the track data packet acquired and uploaded by the equipment end of the user in real time, namely, the single user longitudinally analyzes.
Taking a user population as a target as an example, the target behavior entity description construction method comprises the following steps:
A. and respectively analyzing the data packets of each point in the track section with the preset duration/length from the track data packets of the same time uploaded by each equipment end in real time, wherein the analysis results of the data packets comprise a speed triaxial, an acceleration triaxial and an angular speed triaxial, and vectorizing the analysis results to construct a speed vector, an acceleration triaxial and an angular speed triaxial.
B. And calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector.
And respectively correcting the acceleration triaxial, and respectively calculating the maximum value and the median of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial.
And respectively calculating the maximum value of the first-order difference absolute value and the median of the first-order difference absolute value of each vector in the triaxial vectors of the angular velocity, and calculating the maximum value in the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity.
And constructing a feature set of the target behavior by using the calculation result. The feature set includes a non-zero velocity mean and a first order difference minimum of the velocity vectors, a first order difference absolute value maximum and a first order difference absolute value median of each of the vectors in the acceleration triaxial vectors, a maximum of the first order difference absolute value maximum of each of the angular velocity triaxial vectors, and a first order difference absolute value median of each of the vectors in the angular velocity triaxial vectors.
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to obtain the target behavior entity description. The second preprocessing comprises sub-coding after the first order difference absolute value median of each vector in the non-zero speed average value and the acceleration triaxial vector and the first order difference absolute value median of each angular speed triaxial vector are divided into boxes, constructing a second average value based on the non-zero speed average value and the first order difference minimum value of the speed vector, and respectively carrying out average value processing on the maximum value in the first order difference absolute value maximum value of each angular speed triaxial vector and the first order difference absolute value maximum value of each vector in the acceleration triaxial vector based on corresponding threshold values.
The user population behavior entity description construction process is described in detail below.
A. Track segments with the time interval of S seconds and the length of N are respectively analyzed from track data packets which are uploaded by each equipment end in real time and are at the same time, and speed, acceleration triaxial and angular speed triaxial of each point position data are extracted. A velocity vector V, an acceleration triaxial X, Y, Z, and an angular velocity triaxial H, T, K are formed.
B. Calculating the characteristics of the speed, acceleration and angular velocity dimensions respectively:
calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min 。
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is corrected to obtain X x ,Y y ,Z z . Calculate X x First order difference X of (2) x ' obtain the absolute value of the first order difference |X x ' maximum value DeltaX is calculated x ＝max(|X x ' I). The same applies to find the maximum value delta Y of the first-order difference absolute value of the acceleration of the other two axes y 、ΔZ z . Further calculate the absolute value of the first order difference |X x Median Δx of' | x-median ＝median(|X x ' I), and similarly, find the median delta Y of the first-order difference absolute values of the accelerations of the other two axes y-median 、ΔZ z-median 。
The first-order differential absolute values (H ', (T ', (K ')) are similarly obtained from the angular velocity triaxial vector, the maximum values (delta H, delta T, delta K) of the first-order differential absolute values are obtained, and the maximum value (delta A) of the maximum values of the first-order differential absolute values of the angular velocity triaxial is further obtained g =max (Δh, Δt, Δk), and then the median Δa of the first order difference absolute value is obtained g-median ＝median(|H’|,|T’|,|K’|)。
Final composition of feature set F for constructing user population behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔX x-median ，ΔY y-median ,ΔZ z-median ，ΔA g ，ΔA g-median ]。
C. From feature set F act Constructing entity descriptions O of user population behaviors act ＝[V box ,ΔVF min ，ΔX box ，ΔY box ，ΔZ box ,ΔXF x ，ΔYF y ，ΔZF z ，ΔA box ，ΔAF g ]Wherein:
V box is the velocity mean value V avg Sub-coded vectors after binning. And (3) enabling the speed binning result to have N bins, wherein the sub-coding vector of the speed binning is a vector with the length of N and is initialized to 0. V (V) avg If the bin division result is M number bin, then V box Is marked 1 at the mth position of (c).
ΔX box ，ΔY box ，ΔZ box Respectively delta X x-median ，ΔY y-median ,ΔZ z-median Sub-coded vectors of the binned result.
ΔA box Is delta A g-median Sub-coded vectors of the binned result.
ΔAF g ＝ΔA g /A g0 ，A g0 Is a preset threshold.
ΔXF x＝ ΔX x /G 0 ，G 0 Is a preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 Is a preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 Is a preset threshold.
Taking a single user as a target as an example, the target behavior entity description construction method comprises the following steps:
A. and analyzing the data packet of each point in the track section with the preset time length/length in the track data packet uploaded by the equipment end in real time, wherein the analysis result of the data packet comprises a speed, an acceleration triaxial, an angular speed triaxial and an alarm state, and vectorizing the analysis result to construct a speed vector, an acceleration triaxial, an angular speed triaxial and an alarm vector.
B. And calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector.
And respectively correcting the acceleration triaxial, and respectively calculating the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial.
And respectively calculating the maximum value of the first-order difference absolute value of each vector in the three-axis vector of the angular velocity, and calculating the maximum value of the first-order difference absolute value of each three-axis vector of the angular velocity.
The alarm vector is converted into subcode.
And constructing a feature set of the target behavior by using the calculation result. The feature set comprises a non-zero speed average value and a first-order difference minimum value of the speed vectors, a maximum value of a first-order difference absolute value of each vector in the acceleration triaxial vectors, a maximum value of the first-order difference absolute value maximum value of each angular speed triaxial vector and subcode of the alarm vector.
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to obtain the target behavior entity description. The second preprocessing includes averaging processing of the non-zero velocity average value of the velocity vector, the first order differential absolute value maximum value of each vector in the triaxial vector, and the maximum value of the first order differential absolute value maximum values of each angular velocity triaxial vector based on the corresponding threshold values, respectively, and constructing a second average value based on the non-zero velocity average value and the first order differential minimum value of the velocity vector.
The user driving behavior entity description construction process is described in detail below.
A. And analyzing a track section with the time interval of S seconds and the length of N from the track data packet uploaded by the equipment end in real time, and extracting the speed, the acceleration triaxial, the angular speed triaxial and the alarm state in each point position data. A velocity vector V, an acceleration triaxial vector X, Y, Z, an angular velocity triaxial vector H, T, K, and an alarm vector C are formed.
B. Calculating the characteristics of speed, acceleration, angular velocity and alarm dimension respectively:
calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min 。
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is corrected to obtain X x ,Y y ,Z z . Calculate X x First order difference X of (2) x ' obtain the absolute value of the first order difference |X x ' maximum value DeltaX is calculated x ＝max(|X x ' I). The same applies to find the maximum value delta Y of the first-order difference absolute value of the acceleration of the other two axes y 、ΔZ z 。
The first-order differential absolute values (H ', (T ', (K ')) are similarly obtained from the angular velocity triaxial vector, the maximum values (delta H, delta T, delta K) of the first-order differential absolute values are obtained, and the maximum value (delta A) of the maximum values of the first-order differential absolute values of the angular velocity triaxial is further obtained g ＝max(ΔH，ΔT，ΔK)。
The alarm vector C is converted into subcode Cr.
Final composition of feature set F for constructing user driving behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔA g ，Cr]。
C. From feature set F act Building entity description O of user driving behavior act ＝[VF avg ,ΔVF min ，ΔXF x ，ΔYF y ，ΔZF z ,ΔAF g ，Cr]Wherein:
VF avg ＝V avg /V 0 ，V 0 is a preset threshold.
ΔAF g ＝ΔA g /A g0 ，A g0 Is a preset threshold.
ΔXF x＝ ΔX x /G 0 ，G 0 Is a preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 Is a preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 Is a preset threshold.
Example two
The embodiment discloses a collision detection method based on target behaviors, as shown in fig. 1, which comprises the following steps:
constructing a current target behavior entity description; when the historical target behavior entity description set exists, similarity between the current target behavior entity description and each historical target behavior entity description in the historical target behavior entity description set is calculated respectively, a plurality of historical target behavior entity descriptions with the front similarity are obtained, statistics of the plurality of historical target behavior entity descriptions are extracted, and collision detection is carried out based on the current target behavior entity description and the statistics. And when the historical target behavior entity description set does not exist, performing collision detection based on the current target behavior entity description. The construction method of the current target behavioral entity description comprises the following A, B, C processes:
A. and analyzing the data packets of each point in the track section with the preset duration/length from the track data packets uploaded by the equipment end in real time, and respectively carrying out vectorization processing on the analysis results with specific dimensions.
B. Carrying out first preprocessing on the vectorized analysis result to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension.
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to construct the target behavior entity description.
When there is a historical target behavioral entity description set, collision detection is divided into two phases: and performing first-stage collision detection by a first method based on the current target behavior entity description and statistics, performing second-stage collision detection by a second method based on the current target behavior entity description, and performing overall collision detection based on the results of the first-stage collision detection and the second-stage collision detection. Specifically, the collision detection result is calculated based on the results of the first-stage and second-stage collision detection and the weight of the collision detection in each stage in the overall collision detection.
And when the historical target behavior description set does not exist, directly calculating a collision detection result according to the current target behavior entity description.
The above-mentioned similarity calculation process includes:
for a user population as a target, the feature involved in similarity calculation is a subset of entity description vectors [ V box ，ΔX box ，ΔY box ，ΔZ box ，ΔA box ]Let the current entity description vector subset be O act-current ＝[V box-current ，ΔX box-current ，ΔY box-current ，ΔZ box-current ，ΔA box-current ]One group behavior entity description subset is O act-population ＝[V box-population ，ΔX box-population ，ΔY box-population ，ΔZ box-population ，ΔA box-population ]The distance function between them is:
for a single user as a goal, the feature involved in similarity calculation is the subset of entity-description vectors [ VF ] avg ,ΔVF min ，ΔAF g ，Cr]Let the current entity description vector subset be O act-current ＝[VF avg-current ,ΔVF min-current ，ΔAF g-current ，Cr current ]A historical entity description subset is O act-history ＝[VF avg-history ,ΔVF min-history ，ΔAF g-history ，Cr history ]The distance function between them is:
when there is a historical target behavioral entity description set:
for a user population as a target, the first stage collision detection process based on the current target behavioral entity description and statistics includes:
statistics of each historical target behavioral entity description include Q v ,Q x ，Q y ，Q z ，Q a ，QD v ，QD x ，QD y ，QD z ，QD a Q represents the fractional number and QD represents the fractional distance. The first stage collision detection calculation method comprises the following steps:
n is the total number of track packets of the population, and N is the number of track packets of the current target belonging to the population.
The second stage collision detection process based on the current target behavioral entity description includes:
calculating the probability P of collision of the current behavior current ＝sigmoid(ΔVF min +ΔXF x +ΔYF y +ΔZF z+ ΔAF g )。
The above-mentioned process based on the results of the first-stage and second-stage collision detection includes:
according to P population And P current Calculating the final collision probability p=w 0 *P population +w 1 *P current Wherein w is 0 And w 1 The collision detection of the first stage and the second stage respectively accounts for the weight of the whole collision detection, w 0 +w 1 =1. By default, w 0 ，w 1 Default values for (2) are all 0.5.
For an individual user as a target, the above-described first-stage collision detection process based on current target behavioral entity descriptions and statistics includes:
each historical target rowStatistics described for an entity include Q x ，Q y ，Q z ，QD x ，QD y ，QD z Q represents the fractional number and QD represents the fractional distance. The first stage collision detection calculation method comprises the following steps:
the second stage collision detection process based on the current target behavioral entity description includes:
calculating the probability P of collision of the current behavior current ＝sigmoid(ΔXF x +ΔYF y +ΔZF z )。
The above-mentioned process based on the results of the first-stage and second-stage collision detection includes:
according to P history And P current Calculating the final collision probability p=w 0 *P history +w 1 *P current Wherein w is 0 And w 1 The collision detection of the first stage and the second stage respectively accounts for the weight of the whole collision detection, w 0 +w 1 =1. By default, w 0 ，w 1 Default values for (2) are all 0.5.
When there is no historical target behavioral entity description set:
for a user population as a target, the method for performing collision detection based on the current target behavioral entity description includes:
calculating the probability P of collision of the current behavior current ＝sigmoid(ΔVF min +ΔXF x +ΔYF y +ΔZF z+ ΔAF g )。
For a single user as a target, the method for performing collision detection based on the current target behavior entity description includes:
Calculating the probability P of collision of the current behavior current ＝sigmoid(ΔXF x +ΔYF y +ΔZF z )。
Example III
The embodiment discloses a collision detection method based on user population behaviors, which comprises the following steps:
constructing a current user population behavior entity description; and when the historical user population behavior entity description set exists, respectively calculating the similarity between the current user population behavior entity description and each historical user population behavior entity description in the historical user population behavior entity description set, acquiring a plurality of historical user population behavior entity descriptions with the previous similarity, extracting statistics of the plurality of historical user population behavior entity descriptions, and performing collision detection based on the current user population behavior entity description and the statistics. And when the historical user population behavior entity description set does not exist, collision detection is performed based on the current user population behavior entity description. The construction method of the current user population behavior entity description comprises the following A, B, C steps:
A. and respectively extracting the speed, the acceleration triaxial and the angular velocity triaxial in the point data from each analyzed track data packet corresponding to multiple users, and correspondingly constructing a speed vector, an acceleration triaxial vector and an angular velocity triaxial vector.
B. And calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector.
And respectively correcting the acceleration triaxial, and respectively calculating the maximum value and the median of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial.
And respectively calculating the maximum value of the first-order difference absolute value and the median of the first-order difference absolute value of each vector in the triaxial vectors of the angular velocity, and calculating the maximum value in the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity.
And constructing a characteristic set of the user population behaviors by using the calculation result. The feature set includes a non-zero velocity mean and a first order difference minimum of the velocity vectors, a first order difference absolute value maximum and a first order difference absolute value median of each of the vectors in the acceleration triaxial vectors, a maximum of the first order difference absolute value maximum of each of the angular velocity triaxial vectors, and a first order difference absolute value median of each of the vectors in the angular velocity triaxial vectors.
C. Preprocessing each dimension parameter in the feature set to construct the user population behavior entity description. The preprocessing process comprises the steps of respectively carrying out sub-coding on a non-zero speed average value, a first-order difference absolute value median of each vector in the acceleration triaxial vector and a first-order difference absolute value median of each angular speed triaxial vector after binning, constructing a second average value based on the non-zero speed average value and a first-order difference minimum value of the speed vector, and respectively carrying out average value processing on the maximum value of the first-order difference absolute value maximum value of each angular speed triaxial vector and the first-order difference absolute value maximum value of each vector in the acceleration triaxial vector based on a corresponding threshold value.
In the presence of a historical user population behavioral entity description set, collision detection is divided into two phases: and performing first-stage collision detection based on the current user population behavior entity description and statistics, performing second-stage collision detection based on the current user population behavior entity description, and performing overall collision detection based on the results of the first-stage collision detection and the second-stage collision detection. Specifically, the collision detection result is calculated based on the results of the first-stage and second-stage collision detection and the weight of the collision detection in each stage in the overall collision detection.
And when the historical user population behavior description set does not exist, directly calculating a collision detection result according to the current user population behavior entity description.
Example IV
The embodiment discloses a collision detection method based on driving behavior habits of users, which comprises the following steps:
constructing a current user behavior entity description; when the historical user behavior entity description set exists, similarity between the current user behavior entity description and each historical user behavior entity description in the historical user behavior entity description set is calculated respectively, a plurality of historical user behavior entity descriptions with the front similarity are obtained, statistics of the plurality of historical user behavior entity descriptions are extracted, and collision detection is carried out based on the current user behavior entity description and the statistics. And when the historical user behavior entity description set does not exist, performing collision detection based on the current user behavior entity description. The construction method of the current user behavior entity description comprises the following A, B, C steps:
A. And respectively analyzing the data packets of each point in the track section with the preset time length/length in the track data packets uploaded by the equipment end in real time, wherein the analysis results of the data packets comprise a speed, an acceleration triaxial, an angular speed triaxial and an alarm state, and vectorizing the analysis results to construct a speed vector, an acceleration triaxial vector, an angular speed triaxial vector and an alarm vector.
B. And calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector.
And respectively correcting the acceleration triaxial, and respectively calculating the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial.
And respectively calculating the maximum value of the first-order difference absolute value of each vector in the three-axis vector of the angular velocity, and calculating the maximum value of the first-order difference absolute value of each three-axis vector of the angular velocity.
The alarm vector is converted into subcode.
And constructing a feature set of the user behavior by using the calculation result. The feature set comprises a non-zero speed average value and a first-order difference minimum value of the speed vectors, a maximum value of a first-order difference absolute value of each vector in the acceleration triaxial vectors, a maximum value of the first-order difference absolute value maximum value of each angular speed triaxial vector and subcode of the alarm vector.
C. And carrying out second preprocessing on the parameters of each dimension in the feature set to obtain the user behavior entity description. The second preprocessing includes averaging processing of the non-zero velocity average value of the velocity vector, the first order differential absolute value maximum value of each vector in the triaxial vector, and the maximum value of the first order differential absolute value maximum values of each angular velocity triaxial vector based on the corresponding threshold values, respectively, and constructing a second average value based on the non-zero velocity average value and the first order differential minimum value of the velocity vector.
When there is a historical set of user behavioral entity descriptions, collision detection is split into two phases: and performing first-stage collision detection based on the current target behavior entity description and statistics, performing second-stage collision detection based on the current target behavior entity description, and performing overall collision detection based on the results of the first-stage collision detection and the second-stage collision detection. Specifically, the collision detection result is calculated based on the results of the first-stage and second-stage collision detection and the weight of the collision detection in each stage in the overall collision detection.
And when the historical user behavior description set does not exist, directly calculating a collision detection result according to the current target behavior entity description.
Example five
The embodiment discloses a collision detection method based on user population behaviors, which comprises the following steps:
s1: constructing a current user population behavior entity description:
A. and respectively analyzing track segments with the time interval of S seconds and the length of N from track data packets with the same time uploaded by each equipment end in real time, and extracting the speed, the acceleration triaxial and the angular speed triaxial in each point position data. A velocity vector V, an acceleration triaxial X, Y, Z, and an angular velocity triaxial H, T, K are formed.
B. Calculating the characteristics of the speed, acceleration and angular velocity dimensions respectively:
calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min 。
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is corrected to obtain X x ,Y y ,Z z . Calculate X x First order difference X of (2) x ' obtain the absolute value of the first order difference |X x ' maximum value DeltaX is calculated x ＝max(|X x ' I). The same applies to find the maximum value delta Y of the first-order difference absolute value of the acceleration of the other two axes y 、ΔZ z . Further calculate the absolute value of the first order difference |X x Median Δx of' | x-median ＝median(|X x ' I), and similarly, find the median delta Y of the first-order difference absolute values of the accelerations of the other two axes y-median 、ΔZ z-median 。
From the triaxial vector of angular velocity, the same appliesTaking the first-order difference absolute values |H ', |T ', |K ' g =max (Δh, Δt, Δk), and then the median Δa of the first order difference absolute value is obtained g-median ＝median(|H’|,|T’|,|K’|)。
Final composition of feature set F for constructing user population behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔX x-median ，ΔY y-median ,ΔZ z-median ，ΔA g ，ΔA g-median ]。
C. From feature set F act Constructing entity descriptions O of user population behaviors act ＝[V box ,ΔVF min ，ΔX box ，ΔY box ，ΔZ box ,ΔXF x ，ΔYF y ，ΔZF z ，ΔA box ，ΔAF g ]Wherein:
V box is the velocity mean value V avg Sub-coded vectors after binning. And (3) enabling the speed binning result to have N bins, wherein the sub-coding vector of the speed binning is a vector with the length of N and is initialized to 0. V (V) avg If the bin division result is M number bin, then V box Is marked 1 at the mth position of (c).
ΔX box ，ΔY box ，ΔZ box Respectively delta X x-median ，ΔY y-median ,ΔZ z-median Sub-coded vectors of the binned result.
ΔA box Is delta A g-median Sub-coded vectors of the binned result.
ΔAF g ＝ΔA g /A g0 ，A g0 Is a preset threshold.
ΔXF x＝ ΔX x /G 0 ，G 0 Is a preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 Is a preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 Is a preset threshold.
S2: and acquiring a historical target behavior entity description set. And when the historical target behavior entity description set exists, executing S3, otherwise executing S6.
S3, respectively calculating the similarity between the current user and each type of historical target behavior entity description:
the feature involved in similarity calculation is a subset of entity description vectors [ V box ，ΔX box ，ΔY box ，ΔZ box ，ΔA box ]Let the current entity description vector subset be O act-current ＝[V box-current ，ΔX box-current ，ΔY box-current ，ΔZ box-current ，ΔA box-current ]One group behavior entity description subset is O act-population ＝[V box-population ，ΔX box-population ，ΔY box-population ，ΔZ box-population ，ΔA box-population ]The distance function between them is:
s4: using KNN concept, finding the first K population behaviors most similar to the current (closest to the current) by default k=1, extracting statistics Q of K population behaviors v ,Q x ，Q y ，Q z ，Q a ，QD v ，QD x ，QD y ，QD z ，QD a Q represents the quantile, defaults to 80% quantile, QD represents the quantile distance, the total number of population track packets N, and the number of track packets N that the user belongs to the population. Calculating collision probability of the population:
P population is 0.
S5: calculating the probability P of collision of the current behavior current ＝sigmoid(ΔVF min +ΔXF x +ΔYF y +ΔZF z+ ΔAF g ) The resulting collision occurrence probability p=w 0 *P population +w 1 *P current 。w 0 ，w 1 Is 0.5 and is to satisfy w 0 +w 1 ＝1。
S6, calculating the collision occurrence probability P of the current behavior current ＝sigmoid(ΔVF min +ΔXF x +ΔYF y +ΔZF z+ ΔAF g ) As the final collision probability.
Example six
The embodiment discloses a collision detection method based on user driving behaviors, which comprises the following steps:
s1: constructing a current user population behavior entity description:
A. and analyzing a track section with the time interval of S seconds and the length of N from the track data packet uploaded by the equipment end in real time, and extracting the speed, the acceleration triaxial, the angular speed triaxial and the alarm state in each point position data. A velocity vector V, an acceleration triaxial vector X, Y, Z, an angular velocity triaxial vector H, T, K, and an alarm vector C are formed.
B. Calculating the characteristics of speed, acceleration, angular velocity and alarm dimension respectively:
calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min 。
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is corrected to obtain X x ,Y y ,Z z . Calculate X x First order difference X of (2) x ' obtain the absolute value of the first order difference |X x ' maximum value DeltaX is calculated x ＝max(|X x ' I). The same applies to find the maximum value delta Y of the first-order difference absolute value of the acceleration of the other two axes y 、ΔZ z 。
The first-order difference absolute values (H ', (T ', (K ')) are obtained by the angular velocity triaxial vector in the same way, and the first-order difference is obtainedThe maximum value DeltaH, deltaT, deltaK of the absolute values is divided, and the maximum value DeltaA of the maximum values of the first-order difference absolute values of the three axes of angular velocity is further obtained g ＝max(ΔH，ΔT，ΔK)。
The alarm vector C is converted into subcode Cr.
Final composition of feature set F for constructing user driving behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔA g ，Cr]。
C. From feature set F act Building entity description O of user driving behavior act ＝[VF avg ,ΔVF min ，ΔXF x ，ΔYF y ，ΔZF z ,ΔAF g ，Cr]Wherein:
VF avg ＝V avg /V 0 ，V 0 is a preset threshold.
ΔAF g ＝ΔA g /A g0 ，A g0 Is a preset threshold.
ΔXF x＝ ΔX x /G 0 ，G 0 Is a preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 Is a preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 Is a preset threshold.
S2: and acquiring a historical target behavior entity description set. And when the historical target behavior entity description set exists, executing S3, otherwise executing S6.
S3: and respectively calculating the similarity between the current user and each historical target behavior entity description:
the feature involved in similarity calculation is the entity description vector subset VF avg ,ΔVF min ，ΔAF g ，Cr]Let the current entity description vector subset be O act-current ＝[VF avg-current ,ΔVF min-current ，ΔAF g-current ，Cr current ]A historical entity description subset is O act-history ＝[VF avg-history ,ΔVF min-history ，ΔAF g-history ，Cr history ]。
S4: using KNN concept, find the first K historic driving behaviors most similar to the current (closest to the current), default k=1, extract statistics Q of K historic behaviors x ，Q y ，Q z ，QD x ，QD y ，QD z Q represents the fraction, defaults to 80% fraction, and QD represents the fraction distance. Calculating the probability of collision:
P history is 0.
S5: calculating the probability P of collision of the current behavior current ＝sigmoid(ΔXF x +ΔYF y +ΔZF z ) The resulting collision occurrence probability p=w 0 *P history +w 1 *P current 。w 0 ，w 1 Is 0.5 and is to satisfy w 0 +w 1 ＝1。
S6: calculating the probability P of collision of the current behavior current ＝sigmoid(ΔXF x +ΔYF y +ΔZF z ) As the final collision probability.
Example seven
The embodiment discloses a method for constructing a historical target behavior entity description set, as shown in fig. 2, comprising the following steps:
A. and extracting the track data packet to be clustered of the target.
For the targets of the user population, the track data packet to be clustered is a track data packet with uniform time in the histories of all users. For the targets of a single user, the track data packets to be clustered are all historical track data packets of the user.
And respectively executing B-D for each track data packet:
B. and analyzing the vehicle motion parameters of the preset dimensions from the track data packet, and carrying out vectorization processing on the vehicle motion parameters of each dimension.
For the targets of the user group, the vehicle motion parameters analyzed from the track data packet comprise the speed, the acceleration triaxial and the angular speed triaxial of each point, and the corresponding vectorization processing obtains a speed vector, an acceleration triaxial vector and an angular speed triaxial vector.
For the target of a single user, the vehicle motion parameters analyzed from the track data packet comprise the speed, the acceleration triaxial, the angular speed triaxial and the alarm state of each point, and the corresponding vectorization processing obtains a speed vector, an acceleration triaxial vector, an angular speed triaxial vector and an alarm vector.
C. And carrying out first preprocessing on the vectors of each dimension to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension.
For a target of a user population, the first preprocessing includes:
calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector;
correcting the acceleration triaxial respectively, and calculating the maximum value and the median of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial respectively;
and respectively calculating the maximum value of the first-order difference absolute value and the median of the first-order difference absolute value of each vector in the triaxial vectors of the angular velocity, and calculating the maximum value in the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity.
For the purpose of a single user, the first preprocessing procedure includes:
calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector;
Correcting the acceleration triaxial respectively, and calculating the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial;
respectively calculating the maximum value of the first-order difference absolute value of each vector in the triaxial vector of the angular velocity, and calculating the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity;
the alarm vector is converted into subcode.
D. And carrying out second preprocessing on the parameters of each dimension in the feature set to construct the target behavior entity description.
For the purpose of the user population, the second preprocessing includes:
and respectively carrying out sub-coding on the non-zero speed average value, the first-order difference absolute value median of each vector in the acceleration triaxial vector and the first-order difference absolute value median of each angular speed triaxial vector after binning, constructing a second average value based on the non-zero speed average value and the first-order difference minimum value of the speed vector, and respectively carrying out average value processing on the maximum value of the first-order difference absolute value maximum value of each angular speed triaxial vector and the first-order difference absolute value maximum value of each vector in the acceleration triaxial vector based on corresponding threshold values.
For the purpose of a single user, the second preprocessing procedure includes:
And respectively carrying out mean processing on the non-zero speed mean value of the speed vector, the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector and the maximum value of the first-order difference absolute value of each angular speed triaxial vector based on the corresponding threshold value, and constructing a second mean value based on the non-zero speed mean value and the first-order difference minimum value of the speed vector.
E. And constructing a similarity matrix according to the similarity between the descriptions of the target behavior entities.
And calculating the similarity distance between every two target behavior entity descriptions according to specific parameters in the target behavior entity descriptions, and constructing a similarity matrix according to each calculated similarity distance.
For the targets of the user population, the parameters involved in the similarity calculation include: the non-zero velocity mean value, the first order difference absolute value median of each vector in the acceleration triaxial vector and the first order difference absolute value median of each angular velocity triaxial vector are sub-coded vectors after binning.
For the purposes of a single user, parameters involved in the similarity calculation include: the non-zero velocity mean value, the maximum value of the first order difference absolute value maximum values of the triaxial vectors of each angular velocity, the mean value based on the corresponding threshold, and the second mean value, the alarm vector subcode.
F. And clustering the target behaviors according to the similarity matrix.
Clustering the target behaviors by using a kmeans++ algorithm according to the similarity matrix, and searching parameters through grids to obtain target behavior class clusters of a plurality of classes with optimal targets.
G. Calculating the central point and the related statistic of each target behavior cluster, and storing the central point and the related statistic of each target behavior cluster in a correlated way to obtain a historical target behavior entity description set.
And calculating the central point and related statistics of each target behavior cluster, and constructing a corresponding target behavior cluster vector. And writing each target behavior cluster vector into a target behavior database for use in collision detection.
Example eight
The embodiment discloses a user population behavior clustering method, as shown in fig. 2, comprising the following steps:
s001, extracting track data packets of all users at the same time from a historical database.
S002, analyzing a track segment with the time interval of S seconds and the length of N from each track data packet, and extracting the speed, the acceleration triaxial and the angular velocity triaxial of each point location data. A velocity vector V, an acceleration triaxial X, Y, Z, and an angular velocity triaxial H, T, K are formed.
And S003, calculating corresponding data characteristics according to each vector.
Calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is correctedObtaining X x ,Y y ,Z z . Calculate X x First order difference |X of (1) x ' I, obtaining the absolute value of the first-order difference, and obtaining the maximum value delta X x ＝max(|X x ' I). Delta Y is obtained by the same method y ，ΔZ z . Further calculate the absolute value of the first order difference |X x Median Δx of' | x-median ＝median(|X x ' I), and the same theory finds ΔY y-median ,ΔZ z-median 。
The first-order differential absolute values (H ', (T ', (K ')) are similarly obtained from the angular velocity triaxial vector, the maximum values (delta H, delta T, delta K) of the first-order differential absolute values are obtained, and the maximum value (delta A) of the angular velocity triaxial is further obtained g =max (Δh, Δt, Δk), and then the median Δa of the first order difference absolute value is obtained g-median ＝median(|H’|,|T’|,|K’|)。
Final composition of feature set F for constructing user population behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔX x-median ，ΔY y-median ,ΔZ z-median ，ΔA g ，ΔA g-median ]。
S004, from the feature set F act Constructing entity descriptions O of user population behaviors act ＝[V box ,ΔVF min ，ΔX box ，ΔY box ，ΔZ box ,ΔXF x ，ΔYF y ，ΔZF z ，ΔA box ，ΔAF g ]Wherein:
V box is the velocity mean value V avg Sub-coded vectors after binning. And (3) enabling the speed binning result to have N bins, wherein the sub-coding vector of the speed binning is a vector with the length of N and is initialized to 0. V (V) avg If the bin division result is M number bin, then V box Is marked 1 at the mth position of (c).
ΔVF min ＝(V avg -ΔV min )/V avg If V avg ＝0ΔVF min ＝-1。
ΔX box ，ΔY box ，ΔZ box Respectively delta X x-median ，ΔY y-median ,ΔZ z-median Sub-coded vectors of the binned result.
ΔA box Is delta A g-median Sub-coded vectors of the binned result.
ΔAF g ＝ΔA g /A g0 ，A g0 For a preset threshold, the default value is 180 0 /s。
ΔXF x＝ ΔX x /G 0 ，G 0 The default value is 980mg for the preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 The default value is 980mg for the preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 The default value is 980mg for the preset threshold.
S005, calculating the driving behavior similarity between every two track data packets of all users according to the user-defined population behavior similarity function, and generating a user population behavior similarity matrix.
The feature involved in similarity calculation is a subset of entity description vectors [ V box ，ΔX box ，ΔY box ，ΔZ box ，ΔA box ]Let the subset of the population entity description vectors of a track packet be O act-m ＝[V box-m ，ΔX box-m ，ΔY box-m ，ΔZ box-m ，ΔA box-m ]The population entity description subset of one track packet is O act-n ＝[V box-n ，ΔX box-n ，ΔY box-n ，ΔZ box-n ，ΔA box-n ]。
The distance between them is:
finally, the population behavior similarity matrix of the N track data packets of all the users is obtained through calculation:
s006, clustering the population behaviors of the user by using a kmeans++ algorithm according to the population behavior similarity matrix, and searching parameters through grids to obtain the optimal M-class population behavior clusters of the user.
S007, calculating the center point and the related statistics of each group behavior cluster, forming a group behavior cluster vector c= [ CID, count,<terminal，icount>,CV box ，ΔCX box ，ΔCY box ，ΔCZ box ，ΔCA box ，Q v ,Q x ，Q y ，Q z ，Q a ，QD v ，QD x ，QD y ，QD z ，QD a ]。
CID is the group category identification.
count is the total number of the group track packets
< terminal, icount > is the binary group of N devices in the population < device number, the total number of track packets of the device >
CV box V in the description of all entities for group behavior belonging to the category box Bitwise and operation result of the vector.
ΔCX box DeltaX in all entity descriptions for group behaviors belonging to that class box Bitwise and operation result of the vector.
ΔCY box ΔY in all entity descriptions for population behavior belonging to that class box Bitwise and operation result of the vector.
ΔCZ box ΔZ in all entity descriptions for population behavior belonging to that class box Bitwise and operation result of the vector.
ΔCA box ΔA in all entity descriptions for population behavior belonging to that class box Bitwise and operation result of the vector.
Q v DeltaVF in all entity descriptions for population behavior belonging to that category min Default to 80% quantiles.
Q x ΔFX in all entity descriptions for group behaviors belonging to the category x Default to 80% quantiles.
Q y ΔFY in all entity descriptions for population behavior belonging to that class y Default to 80% quantiles.
Q z ΔFZ in all entity descriptions for population behavior belonging to that class z Default to 80% quantiles.
Q a ΔAF in all entity descriptions for population behavior belonging to that class g Default to 80% quantiles.
QD v DeltaVF in all entity descriptions for population behavior belonging to that category min By default, 75% quantiles-25% quantiles are taken.
QD x ΔFX in all entity descriptions for group behaviors belonging to the category x By default, 75% quantiles-25% quantiles are taken.
QD y ΔFY in all entity descriptions for historic behavior belonging to the category y By default, 75% quantiles to 25% quantiles are taken.
QD z ΔFZ in all entity descriptions for historic behavior belonging to the category z By default, 75% quantiles to 25% quantiles are taken.
QD a ΔAF in all entity descriptions for population behavior belonging to that class g By default, 75% quantiles-25% quantiles are taken.
And S008, writing the result of the S007 into a user population behavior database.
Example nine
The embodiment discloses a user driving behavior clustering method, which comprises the following steps:
s001, extracting all track data packets of a single user from a historical database.
S002, analyzing a track section with the time interval of S seconds and the length of N from each track data packet, and extracting the speed, the acceleration triaxial, the angular velocity triaxial and the alarm state in each point location data. A velocity vector V, an acceleration triaxial vector X, Y, Z, an angular velocity triaxial vector H, T, K, and an alarm vector C are formed.
S003, calculating the data characteristics corresponding to each track data packet.
Calculating the mean value V of non-0 velocity from the velocity vector V avg First order difference minimum DeltaV min
According to the triaxial acceleration coordinate system component correction method, X, Y, Z is corrected to obtain X x ,Y y ,Z z . Calculate X x First order difference X of (2) x ' obtaining the maximum value DeltaX of the absolute value of the first order difference x ＝max(|X x ' I). By the same thing, ΔY y ，ΔZ z 。
The first order difference is obtained from the three axial vectors of the angular velocity, the maximum values DeltaH, deltaT, deltaK of the absolute values of the first order difference are obtained, and the maximum value DeltaA of the three axial vectors of the angular velocity is further obtained g ＝max(ΔH，ΔT，ΔK)。
The alarm vector C is converted into subcode Cr.
Final composition of feature set F for constructing user driving behavior act ＝[V avg ,ΔV min ，ΔX x ，ΔY y ，ΔZ z ,ΔA g ，Cr]。
S004, from the feature set F act Building an entity description O of the driving behavior of each track packet of the user history act ＝[VF avg ,ΔVF min ，ΔXF x ，ΔYF y ，ΔZF z ,ΔAF g ，Cr]Wherein:
VF avg ＝V avg /V 0 ，V 0 the default value is 15km/h for a preset threshold.
ΔVF min ＝(V avg -ΔV min )/V avg If V avg ＝0ΔVF min ＝-1。
ΔAF g ＝ΔA g /A g0 ，A g0 For a preset threshold, the default value is 180 0 /s。
ΔXF x＝ ΔX x /G 0 ，G 0 The default value is 980mg for the preset threshold.
ΔYF y＝ ΔY y /G 0 ，G 0 The default value is 980mg for the preset threshold.
ΔZF z＝ ΔZ z /G 0 ，G 0 The default value is 980mg for the preset threshold.
S005, calculating the driving behavior similarity between every two track data packets of the user history according to the self-defined driving behavior similarity function, and generating a user driving behavior similarity matrix.
The feature involved in similarity calculation is the entity description vector subset VF avg ,ΔVF min ，ΔAF g ，Cr]Let entity description vector subset of a track data packet be O m ＝[VF avg-m ,ΔVF min-m ，ΔAF g-m ，Cr m ]The entity description vector subset of the other track data packet is O n ＝[VF avg-n ,ΔVF min-n ，ΔAF g-n ，Cr n ]The distance between them is:
finally, the historical behavior similarity matrix of the single user with N historical track data packets is obtained by calculation:
s006, clustering the historical behaviors of the user by using a kmeans++ algorithm according to the historical behavior similarity matrix, and searching parameters through grids to obtain the optimal M types of historical behavior clusters of the user.
S007, calculating the central point and the related statistics of each historical behavior cluster to form a warehouse-in historical behavior cluster vector C= [ Terminal, CID, CVF avg ,ΔCVF min ，ΔCAF g ，CCr，Q x ，Q y ，Q z ，QD x ，QD y ，QD z ]
Terminal is the device number.
CID is the device history behavior class identification.
CVF avg VF in a description of historical behavioral entities belonging to that category avg Is a mean value of (c).
ΔCVF min For DeltaVF in a historical behavioural entity description belonging to that category min Is a mean value of (c).
ΔCAF g ΔAF in the entity description for historic behaviors belonging to the category g Is a mean value of (c).
CCr is the result of bitwise AND of Cr in the historical behavioral entity description belonging to that category.
Q x ΔFX in the description of historical behavioral entities belonging to that category x Default to 80% quantiles.
Q y For ΔFY in historical behavioral entity descriptions belonging to that category y Default to 80% quantiles.
Q z For ΔFZ in historical behavioral entity descriptions belonging to that category z Default to 80% quantiles.
QD x ΔFX in the description of historical behavioral entities belonging to that category x By default, 75% quantiles-25% quantiles are taken.
QD y For ΔFY in historical behavioral entity descriptions belonging to that category y By default, 75% quantiles to 25% quantiles are taken.
QD z For ΔFZ in historical behavioral entity descriptions belonging to that category z By default, 75% quantiles to 25% quantiles are taken.
And S008, writing the result of the S007 into a user history behavior database.
The invention is not limited to the specific embodiments described above. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification, as well as to any novel one, or any novel combination, of the steps of the method or process disclosed.
Claims (7)
1. A collision detection method based on target behavior, comprising the steps of:
constructing a current target behavior entity description, and acquiring a historical target behavior entity description set;
when a historical target behavior entity description set exists, similarity between the current target behavior entity description and each historical target behavior entity description in the historical target behavior entity description set is calculated respectively, a plurality of historical target behavior entity descriptions with the front similarity are obtained, statistics of the plurality of historical target behavior entity descriptions are extracted, and collision detection is carried out based on the current target behavior entity description and the statistics;
When the historical target behavior entity description set does not exist, collision detection is carried out based on the current target behavior entity description;
the construction method of the current target behavioral entity description comprises the following A, B, C processes:
A. analyzing vehicle motion parameters of preset dimensions in the track data packet of preset duration/length from the track data packet uploaded by the equipment end in real time, and vectorizing the vehicle motion parameters of each dimension;
B. carrying out first preprocessing on the vectorized analysis result to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension;
C. performing second preprocessing on each dimension parameter in the feature set to construct a target behavior entity description;
the construction method of the historical target behavior entity description set comprises the following steps:
a. extracting a track data packet to be clustered of a target;
b-d is respectively executed for each track data packet:
b. analyzing vehicle motion parameters of preset dimensions from each point of the track data packet, and carrying out vectorization processing on the vehicle motion parameters of each dimension;
c. carrying out first preprocessing on the vectors of each dimension to respectively obtain the characteristics of each dimension, and constructing a characteristic set of the target behavior according to the characteristics of each dimension;
d. Performing second preprocessing on each dimension parameter in the feature set to construct a target behavior entity description;
e. constructing a similarity matrix according to the similarity between the descriptions of each target behavior entity;
f. clustering the target behaviors according to the similarity matrix;
g. calculating the central point and the related statistic of each target behavior cluster, and carrying out association storage on the central point and the related statistic of each target behavior cluster;
in the construction process of the current and/or historical target behavior entity description, vehicle motion parameters analyzed from the track data packet comprise:
for targets of user groups, vehicle motion parameters analyzed from the track data packet comprise a speed, an acceleration triaxial and an angular velocity triaxial, and corresponding vectorization processing is carried out to obtain a speed vector, an acceleration triaxial vector and an angular velocity triaxial vector;
for the targets of a single user, vehicle motion parameters analyzed from the track data packet comprise speed, acceleration triaxial, angular speed triaxial and alarm states, and corresponding vectorization processing is carried out to obtain a speed vector, an acceleration triaxial vector, an angular speed triaxial vector and an alarm vector;
the first preprocessing includes:
for a target of a user population, the first preprocessing includes:
Calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector;
correcting the acceleration triaxial respectively, and calculating the maximum value and the median of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial respectively;
respectively calculating the maximum value of the first-order difference absolute value and the median of the first-order difference absolute value of each vector in the triaxial vector of the angular velocity, and calculating the maximum value in the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity;
for the purpose of a single user, the first preprocessing procedure includes:
calculating a non-zero speed average value and a first-order difference minimum value according to the speed vector;
correcting the acceleration triaxial respectively, and calculating the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector based on the corrected acceleration triaxial;
respectively calculating the maximum value of the first-order difference absolute value of each vector in the triaxial vector of the angular velocity, and calculating the maximum value of the first-order difference absolute value of each triaxial vector of the angular velocity;
the alarm vector is converted into subcode.
2. The target behavior-based collision detection method of claim 1, wherein the second preprocessing includes:
For the purpose of the user population, the second preprocessing includes:
respectively carrying out sub-coding on the non-zero speed average value, the first-order difference absolute value median of each vector in the acceleration triaxial vector and the first-order difference absolute value median of each angular speed triaxial vector after binning, constructing a second average value based on the non-zero speed average value and the first-order difference minimum value of the speed vector, and respectively carrying out average value processing on the maximum value in the first-order difference absolute value maximum value of each angular speed triaxial vector and the first-order difference absolute value maximum value of each vector in the acceleration triaxial vector based on corresponding threshold values;
for the purpose of a single user, the second preprocessing procedure includes:
and respectively carrying out mean processing on the non-zero speed mean value of the speed vector, the maximum value of the first-order difference absolute value of each vector in the acceleration triaxial vector and the maximum value of the first-order difference absolute value of each angular speed triaxial vector based on the corresponding threshold value, and constructing a second mean value based on the non-zero speed mean value and the first-order difference minimum value of the speed vector.
3. The target behavior-based collision detection method as claimed in claim 2, wherein the process of collision detection based on the current target behavior entity description and statistics comprises:
And performing first-stage collision detection by a first method based on the current target behavior entity description and statistics, performing second-stage collision detection by a second method based on the current target behavior entity description, and performing overall collision detection based on the results of the first-stage collision detection and the second-stage collision detection.
4. The method for collision detection based on target behavior according to claim 3, wherein the method for overall collision detection based on the results of the first-stage and second-stage collision detection comprises: and calculating the overall collision detection result according to the weight of the overall collision detection occupied by the first-stage collision detection and the second-stage collision detection and combining the settlement results of the first-stage collision detection and the second-stage collision detection.
5. The method for detecting collision based on target behavior according to claim 3 or 4, wherein in the first stage of collision detection based on the current target behavior entity description and statistics in a first method, for the case that the target is a user population, the first method is:
wherein, ppotential is the first stage collision detection result, N is the total number of track packets of the population, N users belong to the track packet number of the population,
，V avg is the non-0 velocity mean of the velocity vector, deltaV min ΔAF, the first order difference minimum of the velocity vector g = ΔA g / A g0 ，A g0 To preset threshold, deltaXF x = ΔX x /G0，ΔYF y = ΔY y /G0，ΔZF z = ΔZ z /G0 ，ΔX x 、ΔY y 、ΔZ z The maximum value of the first-order difference absolute values of the acceleration triaxial is respectively G0 is a preset threshold value, and delta A g For the maximum value of the maximum values of the first-order difference absolute values of the three axes of angular velocity, K is the number of the obtained historical target behavior entity descriptions during similarity comparison, Q represents the quantile number of the obtained K historical target behavior entity descriptions, and QD represents the quantile distance;
for the case of targeting a single user, the first method is:
P history for the first stage collision detection result, Δxf x = ΔX x /G0，ΔYF y = ΔY y /G0，ΔZF z = ΔZ z /G0 ，ΔX x 、ΔY y 、ΔZ z The method comprises the steps of respectively obtaining the maximum value of first-order difference absolute values of three axes of acceleration, wherein G0 is a preset threshold, K is the number of historical target behavior entity descriptions obtained during similarity comparison, Q represents the quantile of the obtained K historical target behavior entity descriptions, and QD represents the quantile distance.
6. The method for detecting collision based on target behavior according to claim 2, wherein the parameters involved in the similarity calculation in the process of calculating the similarity between the current target behavior entity description and each of the historical target behavior entity descriptions in the set of historical target behavior entity descriptions respectively include:
for the case where the goal is a user population, the parameters involved in the similarity calculation include: the non-zero velocity mean value, the first-order difference absolute value median of each vector in the acceleration triaxial vector and the first-order difference absolute value median of each angular velocity triaxial vector are sub-coded vectors after binning;
For the case of a single user being targeted, the parameters involved in the similarity calculation include: the non-zero velocity mean value, the maximum value of the first order difference absolute value maximum values of the triaxial vectors of each angular velocity, the mean value based on the corresponding threshold, and the second mean value, the alarm vector subcode.
7. The collision detection method as claimed in claim 6, wherein the similarity calculating process is a process of calculating a similarity distance between parameter vectors participating in calculation in two entity descriptions, and the similarity distance calculating method is as follows:
for the targets of the user population, the similarity distance calculation method comprises the following steps:
wherein O is act-current 、O act-population Respectively two compared target entity description subsets, wherein the target entity description subsets are composed of parameters participating in similarity calculation;
for the targets of a single user, the similarity distance calculation method is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010316688.XA CN111506692B (en) | 2020-04-21 | 2020-04-21 | Collision detection method based on target behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010316688.XA CN111506692B (en) | 2020-04-21 | 2020-04-21 | Collision detection method based on target behaviors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111506692A CN111506692A (en) | 2020-08-07 |
| CN111506692B true CN111506692B (en) | 2023-05-26 |
Family
ID=71872978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010316688.XA Active CN111506692B (en) | 2020-04-21 | 2020-04-21 | Collision detection method based on target behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111506692B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112231389B (en) * | 2020-10-16 | 2021-08-24 | 中国民用航空华东地区空中交通管理局 | Track-based visual conflict model construction method and device, electronic equipment and storage medium |
| CN114330449B (en) * | 2021-12-31 | 2025-05-27 | 成都路行通信息技术有限公司 | A vehicle collision detection method and system based on feature time domain matching |
| CN114998847B (en) * | 2022-05-10 | 2025-10-24 | 杭州阿里云飞天信息技术有限公司 | Traffic incident detection method, image detection method and computing device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104882025A (en) * | 2015-05-13 | 2015-09-02 | 东华大学 | Crashing detecting and warning method based on vehicle network technology |
| ITUA20162009A1 (en) * | 2016-03-24 | 2017-09-24 | Viasat Group S P A | EXPERT SYSTEM FOR THE INTEGRATED MANAGEMENT OF SAFETY OF DRIVING A PLURALITY OF VEHICLES AND THE DETERMINATION OF AN INTERACTIVE MAP OF THE ROAD RISK |
| CN107945574A (en) * | 2017-10-25 | 2018-04-20 | 东软集团股份有限公司 | A kind of vehicle collision prewarning method, device and equipment |
| CN108010388A (en) * | 2018-01-04 | 2018-05-08 | 北京瑞腾中天科技有限公司 | Collision detection method for early warning and collision detection early warning system based on car networking network |
| US10235882B1 (en) * | 2018-03-19 | 2019-03-19 | Derq Inc. | Early warning and collision avoidance |
| CN109649386A (en) * | 2018-12-26 | 2019-04-19 | 成都路行通信息技术有限公司 | A kind of vehicle collision detection method based on machine learning |
| CN109741483A (en) * | 2018-12-11 | 2019-05-10 | 成都路行通信息技术有限公司 | A car collision detection method based on the Internet of Vehicles platform |
| CN109849904A (en) * | 2019-03-27 | 2019-06-07 | 重庆卡佐科技有限公司 | Vehicle collision identification device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9392431B2 (en) * | 2014-09-30 | 2016-07-12 | Verizon Patent And Licensing Inc. | Automatic vehicle crash detection using onboard devices |
-
2020
- 2020-04-21 CN CN202010316688.XA patent/CN111506692B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104882025A (en) * | 2015-05-13 | 2015-09-02 | 东华大学 | Crashing detecting and warning method based on vehicle network technology |
| ITUA20162009A1 (en) * | 2016-03-24 | 2017-09-24 | Viasat Group S P A | EXPERT SYSTEM FOR THE INTEGRATED MANAGEMENT OF SAFETY OF DRIVING A PLURALITY OF VEHICLES AND THE DETERMINATION OF AN INTERACTIVE MAP OF THE ROAD RISK |
| CN107945574A (en) * | 2017-10-25 | 2018-04-20 | 东软集团股份有限公司 | A kind of vehicle collision prewarning method, device and equipment |
| CN108010388A (en) * | 2018-01-04 | 2018-05-08 | 北京瑞腾中天科技有限公司 | Collision detection method for early warning and collision detection early warning system based on car networking network |
| US10235882B1 (en) * | 2018-03-19 | 2019-03-19 | Derq Inc. | Early warning and collision avoidance |
| CN109741483A (en) * | 2018-12-11 | 2019-05-10 | 成都路行通信息技术有限公司 | A car collision detection method based on the Internet of Vehicles platform |
| CN109649386A (en) * | 2018-12-26 | 2019-04-19 | 成都路行通信息技术有限公司 | A kind of vehicle collision detection method based on machine learning |
| CN109849904A (en) * | 2019-03-27 | 2019-06-07 | 重庆卡佐科技有限公司 | Vehicle collision identification device |
Non-Patent Citations (2)
| Title |
|---|
| Yunpeng Wang.Vehicle collision warning system and collision dectection algorithm based on vehicle infrastructure integration.《7th Advanced Forum on Transpotation of China》.2012,全文. * |
| 蒋红亮.基于深度学习的共享电动汽车碰撞检测与预警研究.《中国优秀硕士学位论文全文数据库工程科技II辑》.2019,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111506692A (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111506692B (en) | Collision detection method based on target behaviors | |
| CN111460443A (en) | Security defense method for data manipulation attack in federated learning | |
| Tran et al. | Improved PSO for feature selection on high-dimensional datasets | |
| CN111211994A (en) | Network traffic classification method based on SOM and K-means fusion algorithm | |
| Cassidy et al. | Calculating feature importance in data streams with concept drift using Online Random Forest | |
| CN113627075B (en) | Projectile pneumatic coefficient identification method based on adaptive particle swarm optimization extreme learning | |
| CN103901880A (en) | Industrial process fault detection method based on multiple classifiers and D-S evidence fusion | |
| CN111506627B (en) | Target behavior clustering method and system | |
| CN113257364A (en) | Single cell transcriptome sequencing data clustering method and system based on multi-objective evolution | |
| Mousavirad et al. | Feature selection using modified imperialist competitive algorithm | |
| Pirlo et al. | Fuzzy-zoning-based classification for handwritten characters | |
| Shafronenko et al. | Fuzzy clustering of distorted observations based on optimal expansion using partial distances | |
| Choong et al. | Modeling of vehicle trajectory using K-means and fuzzy C-means clustering | |
| Ratanavilisagul | A novel modified particle swarm optimization algorithm with mutation for data clustering problem | |
| CN114897109A (en) | Mower abnormity monitoring and early warning method | |
| Himmelspach et al. | Clustering approaches for data with missing values: Comparison and evaluation | |
| Da Silva et al. | Validity index-based vigilance test in adaptive resonance theory neural networks | |
| Thanh et al. | An approach to reduce data dimension in building effective network intrusion detection systems | |
| Khushaba et al. | Feature subset selection using differential evolution | |
| CN113765921A (en) | Abnormal flow grading detection method for industrial Internet of things | |
| CN114090967A (en) | A PSO-MSVM-based APT tissue tracking and traceability method and system | |
| CN108446858B (en) | Air-to-ground threat assessment method based on specific network structure | |
| Ratanavilisagul | Modification fitness function of particle swarm optimization to improve the cluster centroid | |
| Seguí et al. | Weighted bagging for graph based one-class classifiers | |
| CN110796237B (en) | Method and device for detecting attack resistance of deep neural network |
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 |