CN115797403A - Traffic accident prediction method and device, storage medium and electronic device - Google Patents
Traffic accident prediction method and device, storage medium and electronic device Download PDFInfo
- Publication number
- CN115797403A CN115797403A CN202211501436.XA CN202211501436A CN115797403A CN 115797403 A CN115797403 A CN 115797403A CN 202211501436 A CN202211501436 A CN 202211501436A CN 115797403 A CN115797403 A CN 115797403A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- data
- traffic accident
- target
- track
- 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.)
- Pending
Links
Images
Classifications
-
- 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
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic accident prediction method and device, a storage medium and electronic equipment. Wherein the method comprises the following steps: acquiring vehicle track data corresponding to a target vehicle set in a preset time period, and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model; and inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident. The invention solves the technical problem of lower accuracy in identifying and predicting the traffic accident in the related technology.
Description
Technical Field
The invention relates to the technical field of traffic information processing, in particular to a traffic accident prediction method and device, a storage medium and electronic equipment.
Background
The current social traffic volume is sharply increased, and traffic accidents often happen, so that long-time traffic jam is caused, and the trip plan of a user is influenced. For the current and future intelligent traffic times, how to quickly find and broadcast accidents is a big problem. The existing accident determination method has few scenes which can be identified and predicted, and the actual road traffic condition has extremely strong complexity and uncertainty, so that the related technology still has the problem of low accuracy when identifying and predicting the traffic accident.
Disclosure of Invention
The embodiment of the invention provides a traffic accident prediction method and device, a storage medium and electronic equipment, which are used for at least solving the technical problem that the accuracy of identifying and predicting a traffic accident is low in the related technology.
According to an aspect of an embodiment of the present invention, there is provided a traffic accident prediction method, including: acquiring vehicle track data corresponding to a target vehicle set in a preset time period, and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model; and inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident.
According to another aspect of the embodiments of the present invention, there is also provided a traffic accident prediction apparatus including: the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring vehicle track data corresponding to a target vehicle set in a preset time period and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set; the second obtaining unit is used for obtaining feature data corresponding to a target theme based on the original parking sample; the training unit is used for training a preset gated neural network model according to the characteristic data to obtain a trained traffic accident prediction model; and the determining unit is used for inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model so as to determine whether the current vehicle has a traffic accident.
According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the traffic accident prediction method through the computer program.
According to still another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned traffic accident prediction method when running.
In the embodiment of the invention, vehicle track data corresponding to a target vehicle set in a preset time period are obtained, and the vehicle track data are preprocessed to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model; in the method, the characteristic data of different subject types are obtained according to the target vehicle track, the gated neural network model can be trained according to scene-conforming and refined characteristics to obtain the trained traffic accident prediction model, and the trained traffic accident prediction model is further used for determining whether the current vehicle has the traffic accident or not, so that the accuracy of recognizing and predicting the traffic accident can be improved, and the technical problem that the accuracy of recognizing and predicting the traffic accident is lower in the related technology is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a schematic diagram of an application environment of an alternative traffic accident prediction method according to an embodiment of the invention;
FIG. 2 is a schematic diagram of an application environment of an alternative traffic accident prediction method according to an embodiment of the invention;
FIG. 3 is a flow diagram illustrating an alternative traffic accident prediction method according to an embodiment of the present invention;
FIG. 4 is a schematic view of an alternative traffic accident prediction model according to an embodiment of the present invention;
FIG. 5 is a schematic flow diagram of an alternative traffic accident prediction method according to an embodiment of the present invention;
FIG. 6 is a flow diagram illustrating an alternative traffic accident prediction method according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of an alternative traffic accident prediction apparatus according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an alternative electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Moreover, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to an aspect of the embodiment of the present invention, a traffic accident prediction method is provided, and optionally, as an optional implementation, the traffic accident prediction method may be applied, but not limited, to the application environment as shown in fig. 1. The application environment comprises: the terminal equipment 102, the network 104 and the server 106 are used for human-computer interaction with the user. The user 108 and the terminal device 102 can perform human-computer interaction, and a traffic accident prediction application program runs in the terminal device 102. The terminal device 102 includes a human-machine interaction screen 1022, a processor 1024, and a memory 1026. The human-computer interaction screen 1022 is configured to display vehicle trajectory data corresponding to a set of target vehicles within a preset time period; the processor 1024 is configured to obtain vehicle trajectory data corresponding to a set of target vehicles within a preset time period. The memory 1026 is configured to store vehicle trajectory data corresponding to the target vehicle set within the preset time period.
In addition, the server 106 includes a database 1062 and a processing engine 1064, where the database 1062 is used to store the planned path determined according to the starting point and the ending point input by the user. The processing engine 1064 is configured to obtain vehicle trajectory data corresponding to a target vehicle set in a preset time period, and pre-process the vehicle trajectory data to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained gated neural network model; inputting data to be predicted corresponding to the current vehicle into the trained gated neural network model to determine whether the current vehicle has a traffic accident; and sending the prediction result of whether the traffic accident occurs to the client of the terminal device 102 for displaying.
In one or more embodiments, the traffic accident prediction method described above in the present application may be applied to the application environment shown in fig. 2. As shown in fig. 2, a human-computer interaction may be performed between a user 202 and a user device 204. The user equipment 204 includes a memory 206 and a processor 208. The user device 204 in this embodiment may refer to, but is not limited to, performing the above-described operations performed by the terminal device 102 to obtain and present the prediction result of the traffic accident.
Optionally, the terminal device 102 and the user device 204 include, but are not limited to, a mobile phone, a tablet computer, a notebook computer, a PC, a vehicle-mounted electronic device, a wearable device, and the like, and the network 104 may include, but is not limited to, a wireless network or a wired network. Wherein, this wireless network includes: WIFI and other networks that enable wireless communication. Such wired networks may include, but are not limited to: wide area networks, metropolitan area networks, and local area networks. The server 106 may include, but is not limited to, any hardware device capable of performing computations. The server may be a single server, a server cluster composed of a plurality of servers, or a cloud server. The above is merely an example, and this is not limited in this embodiment.
As an alternative implementation, as shown in fig. 3, an embodiment of the present invention provides a traffic accident prediction method, including the following steps:
s302, vehicle track data corresponding to a target vehicle set in a preset time period are obtained, and the vehicle track data are preprocessed to obtain an original parking sample corresponding to the target vehicle set.
Specifically, in the embodiment of the present invention, the vehicle trajectory data includes, but is not limited to, a vehicle unique identifier (license plate number and vehicle color), timestamps of different track points, longitude of the track point, latitude of the track point, alarm information reported by the vehicle, vehicle speed, acceleration of the vehicle, an included angle between a vehicle head and a north side of the vehicle, information on whether the vehicle turns on a turn signal, a road type (high speed, national road, provincial road, county road, etc.) where the vehicle is located, intersection information, and the like.
S304, acquiring feature data corresponding to the target subject based on the original docking sample.
Specifically, the target subjects include, but are not limited to, target vehicles, surrounding vehicles, and roads, and the original parking samples are classified according to the target subjects, and the related feature information corresponding to each target subject is calculated.
S306, training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model.
Specifically, the data set of the feature data comprises a training set, a verification set and a test set; wherein, the positive sample is the stopping track data of the traffic accident through manual verification as the positive sample; the negative sample is the elimination of accident data from the original parking data. In the training set, positive samples are kept in a certain proportion by resampling with putting back. And (3) extracting vehicle parking track data of several days by using the negative sample, and maintaining the positive sample and the negative sample as a data set with a preset proportion, wherein the data volume of the positive sample is less than that of the negative sample.
And S308, inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident.
Specifically, the data to be predicted corresponding to the current vehicle is input into the trained traffic accident prediction model, the prediction probability of whether the current vehicle has a traffic accident is obtained, for example, the prediction probability of the current vehicle having the traffic accident is 0.8 and is greater than a preset threshold value 0.6, and the prediction label is output as the traffic accident. For another example, the prediction probability of the current occurrence of the traffic accident is 0.5, which is smaller than the preset threshold value of 0.6, and the prediction label is output as the non-occurrence of the traffic accident.
In the embodiment of the invention, vehicle track data corresponding to a target vehicle set in a preset time period are obtained, and the vehicle track data are preprocessed to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model; in the method, the characteristic data of different subject types are obtained according to the target vehicle track, the gated neural network model can be trained according to scene-conforming and refined characteristics to obtain the trained traffic accident prediction model, and the trained traffic accident prediction model is further used for determining whether the current vehicle has the traffic accident or not, so that the accuracy of recognizing and predicting the traffic accident can be improved, and the technical problem that the accuracy of recognizing and predicting the traffic accident is lower in the related technology is solved.
In one or more embodiments, the preprocessing the vehicle trajectory data to obtain an original parking sample corresponding to the target vehicle set includes:
sequencing the vehicle track data according to time;
for each vehicle in the target vehicle set, if the speeds of two adjacent track points are all zero and the time difference between the two adjacent track points is within a preset time range, taking the previous track point in the two adjacent track points as a starting stop point;
if the subsequent stop point of the current vehicle is within a first preset distance from the starting stop point and the duration within the first preset distance exceeds a first preset duration, judging whether the vehicle stops at the interest point or whether the distance between the vehicle and the interest point is within a second preset distance;
and if the current vehicle does not stop at the interest point or the distance between the current vehicle and the interest point is greater than the second preset distance, determining the vehicle track data corresponding to the current vehicle as an original stop sample corresponding to the vehicle.
In one or more embodiments, the obtaining feature data corresponding to a target topic based on the original docking sample includes:
screening out a track data sequence to be subjected to feature calculation with a second preset time length from the original docking sample;
classifying the track data sequence according to a target vehicle, vehicles around the target vehicle and a road where the target vehicle is located to obtain a track data sequence corresponding to each topic type; wherein the target subject comprises a target vehicle, vehicles around the target vehicle and a road where the target vehicle is located;
and determining the characteristic data of the track data sequence corresponding to each topic category.
As shown in fig. 4, the trajectory data series at different times corresponding to a plurality of continuous vehicles with the same interval time are input into the traffic accident prediction model, so that whether the vehicle has a traffic accident can be predicted.
In one or more embodiments, determining feature data of a trajectory data sequence corresponding to a subject category of a target vehicle includes:
whether the current vehicle is in a collision alarm state, whether the current vehicle is in a running state or a parking state, whether the current vehicle is in a double-flash state or not and whether the current target vehicle is in a single-flash state or not are converted into binary-type first preprocessing data through a vehicle-mounted terminal of the target vehicle.
Specifically, the current vehicle collision alarm is converted into 1, the collision alarm is not converted into 0, and the current vehicle double-flash on is converted into 1; when the vehicle is in a running state, the configuration characteristic value is 1, when the vehicle is in a parking state, the configuration characteristic value is 0, and when the vehicle is not opened, the double-flash is converted into 0; the current target vehicle is an open single flash converted to 1, and an unopened vehicle single flash is converted to 0, so the first preprocessed data includes each feature type and the corresponding 0 or 1 equivalent.
And carrying out abnormal value processing, box sealing processing and normalization processing on the current speed and the acceleration of the target vehicle, the time difference between different tracks and the included angle between the vehicle head and the due north direction to obtain second preprocessing data.
In particular, by the formula
Determining a current speed characteristic value of the target vehicle, wherein Vmax is an analysis predicted value characteristic value, and V is an original speed characteristic value; by the formula
To determine an acceleration characteristic value of the target vehicle, wherein a min And a max Are all analysis prediction eigenvalues; by the formula
To determine a time difference characteristic value between different tracks of the current vehicle, wherein t c Time stamp feature value, t, for the current trace point b Time stamp feature values of the previous track point of the current track point; by the formula
Determining the characteristic value of an included angle between the head of the target vehicle and the due north direction, wherein g 1 Is the feature value g of the north angle of the current track point of the vehicle 0 Is the characteristic value of the north angle of the previous track point, g min And g max Are all analysis predicted eigenvalues.
And taking the first preprocessing data and the second preprocessing data as feature data of a track data sequence corresponding to the subject category of the target vehicle.
In one or more embodiments, determining feature data of a trajectory data sequence corresponding to a vehicle theme category around the target vehicle further includes:
determining a current timestamp of the target vehicle, and taking a union of a preset time period before a time corresponding to the timestamp and a preset time period after the time corresponding to the timestamp as a screening time period;
determining a search area by taking the geographic position of the target vehicle as a circle center and a preset distance as a radius based on the screening time period;
determining track points of the surrounding vehicles in the search area, and determining the average speed of each track point as the average speed value of the surrounding vehicles;
and taking the average speed value as the characteristic data of the track data sequence corresponding to the vehicle theme category around the target vehicle.
In particular, by the formula
The speed of the surrounding vehicle is determined, and then the average value of the speeds of the vehicles is used as the characteristic data of the track data sequence corresponding to the vehicle subject categories around the target vehicle.
In one or more embodiments, determining feature data of a track data sequence corresponding to a road theme category in which the target vehicle is located further includes:
and carrying out binarization processing on the type of the road where the target vehicle is located and whether an intersection exists around the target vehicle at the current point to obtain the characteristic data of the track data sequence corresponding to the subject type of the road where the target vehicle is located.
Specifically, the road type here includes national road, provincial road, prefectural road, high speed, and the like, and the road type is subjected to binarization processing by one-hot encoding.
In one or more embodiments, the training a preset gated neural network model according to the feature data corresponding to the target topic to obtain a trained traffic accident prediction model includes:
acquiring a sample track data set; wherein the sample trajectory data set comprises training samples and validation samples;
configuring a verification category label for whether a traffic accident happens or not for each sample track data in the verification sample according to a traffic accident rule;
inputting training track data and corresponding accident judgment category labels in the training samples into an initialized gated neural network model for training to obtain a training output result, wherein in each training process of the gated neural network model, the accident judgment category labels corresponding to the track data are determined according to the number of track point sequences and the number of characteristics corresponding to the training track data;
and under the condition that the training output result indicates that the convergence condition is reached, determining to obtain the gated neural network model, wherein the convergence condition is used for indicating that the difference degree between the determined class label of the accident judgment of the training track data and the verification class label is smaller than or equal to a preset threshold value, and the verification class label is the class label which has the same track point sequence number and the same characteristic number with the training track data in the verification sample.
In one or more embodiments, the inputting data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether a traffic accident occurs to the current vehicle includes:
converting data to be predicted corresponding to the current vehicle into a feature matrix comprising the number of track point sequences and the number of features;
inputting the characteristic matrix serving as input characteristics into a first layer gating network module of the trained gating neural network model to obtain a first output tensor;
inputting the first output result to a second layer gating network module which inputs the trained gating neural network model to obtain a first splicing tensor of the starting time and a second splicing tensor of the ending time;
inputting the first splicing tensor and the second splicing tensor into a full connection layer of the trained gated neural network model to obtain the probability of whether a current vehicle has a traffic accident or not;
and determining whether a traffic accident occurs according to the probability.
Specifically, as shown in fig. 5, input data is sorted into a feature matrix of M = [ L (number of trace point sequences), H (number of features) ] as an input feature, and a label y is binarized as an output label.
The traffic accident prediction model in the embodiment of the invention adopts a classifier network structure of a Gate controlled Recurrent Unit (GRU) as a two-layer bidirectional GRU Recurrent neural network, inputs the input characteristics into a first layer of GRU, and realizes the formula of O 1 ，H n1 = GRU (M), where O is the output all states, H is the output last state, and M is the feature matrix of batch input; the result of the first layer is then input into the second layer to achieve the formula O 2 ，H n2 ＝GRU(O 1 ) The output layer is a two-class fully-connected layer, the input is the tensor output at the first and last moments of the GRU network using the last layer after splicing, and the realization formula is y = Softmax (O) 21 ，O 2n )。
In one or more embodiments, the traffic accident prediction method further includes: and learning the super-parameters of the gated neural network model through a Bayesian parameter-adjusting algorithm, and obtaining the trained traffic accident prediction model when determining to obtain the target super-parameters.
In an application embodiment, as shown in fig. 6, the traffic accident prediction method includes the following steps:
step 1, obtaining data
The method comprises the steps of obtaining track data of vehicles within a certain time range, wherein each record comprises a unique mark (license plate number and vehicle color) of the vehicle, timestamps of different track points, longitude of the track points, latitude of the track points, alarm information reported by the vehicle, vehicle speed, acceleration of the vehicle, an included angle between the head of the vehicle and the north, information whether the vehicle turns on a turn light, road types (high speed, national road, provincial road, county road and the like) where the vehicle is located, intersection information and the like.
Step 2, stop judgment
Analyzing vehicle track data, sequencing each vehicle according to time, judging whether the speed of two adjacent track points is smaller than a certain threshold value, and the time difference of the two adjacent track points is within a certain time range, if the conditions are met, defining the previous stop point as a starting stop point, defining the distance between the next track point and the starting stop point as a certain distance range, and if the continuous total time of the state exceeds a certain time duration, defining the stop point as a possible accident, judging whether the distance between the POI (point of interest) and the starting stop point of the vehicle, such as a service area, a high-speed parking area, a check station, a toll station and the like, meets the threshold value smaller than the certain distance, filtering stop information of the possible accident, and obtaining the reserved data as an original stop sample.
Step 3, characteristic engineering
Firstly, a track sequence which needs to be subjected to feature calculation is filtered out according to a time boundary of the stopping time t minus a certain time t1 and the stopping time plus a certain time t2, then, classification is carried out according to three subjects, namely a target vehicle, a surrounding vehicle and a road, and the relevant features of each subject are calculated.
Target vehicle characteristics:
in particular, by the formula
Determining a current speed characteristic value of the target vehicle, wherein Vmax is an analysis prediction characteristic value, and V is an original speed characteristic value; by the formula
To determine an acceleration characteristic value of the target vehicle, wherein a min And a max All are analysis prediction eigenvalue values; by the formula
To determine a characteristic value of the time difference between different trajectories of the current vehicle, wherein t c Time stamp feature value, t, for the current trace point b Time stamp feature values of the previous track point of the current track point; by the formula
Determining the characteristic value of an included angle between the head of the target vehicle and the due north direction, wherein g 1 Is the feature value g of the north angle of the current track point of the vehicle 0 Is the characteristic value of the north angle of the previous track point, g min And g max Are all analysis predicted eigenvalues.
Peripheral vehicle characteristics:
in particular, by formula
The speeds of the surrounding vehicles are determined, and then the average value of the speeds of the vehicles is used as the characteristic data of the track data sequence corresponding to the vehicle theme category around the target vehicle.
Road characteristics:
and (4) carrying out binarization processing on the type of the road where the current point is located and whether an intersection exists around the current point. Specifically, the road type here includes national road, provincial road, prefectural road, high speed, and the like, and the road type is subjected to binarization processing by one-hot encoding.
Step 4, constructing a sample data set
Positive sample: the parking of the traffic accident is manually verified to be used as a positive sample; negative sample: original vehicle parking data are removed as much as possible, and accident data are used as negative samples. The data set comprises three parts of a training set, a verification set and a test set. The construction process comprises the following steps: training set: the positive samples are held in a certain proportion by the resample with the replace. Negative samples were taken for several days of docking data; and (4) verification set: the positive and negative sample retention ratio is a preset ratio Np: nn.
Step 5, model selection
Training a GRU classifier model:
specifically, as shown in fig. 5, input data is sorted into a feature matrix of M = [ L (number of trace point sequences), H (number of features) ] as an input feature, and a label y is binarized as an output label.
The traffic accident prediction model in the embodiment of the invention adopts a classifier network structure of a Gate controlled Recurrent Unit (GRU) as a two-layer bidirectional GRU Recurrent neural network, inputs the input characteristics into a first layer of GRU, and realizes the formula of O 1 ，H n1 = GRU (M), where O is the output all states, H is the output last state, and M is the feature matrix of batch input; the result of the first layer is then input into the second layer to achieve the formula O 2 ，H n2 ＝GRU(O 1 ) The output layer is a two-class fully-connected layer, the input is the tensor output at the first and last moments of the GRU network using the last layer after splicing, and the realization formula is y = Softmax (O) 21 ，O 2n )。
And learning the hyper-parameters of the model by a Bayesian parameter adjusting method, and exporting the model after learning the optimal parameters to obtain the trained traffic accident prediction model.
Step 6, identifying types
And loading pre-trained GRU classifier model parameters, and sorting the prediction data into an input format meeting the GRU classifier, so that the probability of an unknown data accident can be predicted, and if the probability is greater than a certain threshold value, the traffic accident is determined to be sent.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art will appreciate that the embodiments described in this specification are presently preferred and that no acts or modules are required by the invention.
According to another aspect of the embodiment of the present invention, there is also provided a traffic accident prediction apparatus for implementing the traffic accident prediction method described above. As shown in fig. 7, the apparatus includes:
a first obtaining unit 702, configured to obtain vehicle trajectory data corresponding to a target vehicle set within a preset time period, and pre-process the vehicle trajectory data to obtain an original parking sample corresponding to the target vehicle set;
a second obtaining unit 704, configured to obtain feature data corresponding to a target topic based on the original docking sample;
the training unit 706 is used for training a preset gated neural network model according to the characteristic data to obtain a trained traffic accident prediction model;
the determining unit 708 is configured to input data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether a traffic accident occurs in the current vehicle.
In the embodiment of the invention, vehicle track data corresponding to a target vehicle set in a preset time period are obtained, and the vehicle track data are preprocessed to obtain an original parking sample corresponding to the target vehicle set; acquiring feature data corresponding to a target subject based on the original docking sample; training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model; in the method, because the characteristic data of different theme types are obtained according to the target vehicle track, the gated neural network model can be trained according to scene-conforming and refined characteristics to obtain the trained traffic accident prediction model, and further the trained traffic accident prediction model is used for determining whether the current vehicle has the traffic accident, so that the accuracy of recognizing and predicting the traffic accident can be improved, and the technical problem that the accuracy of recognizing and predicting the traffic accident is lower in the related technology is solved.
In one or more embodiments, the first obtaining unit 702 includes:
the sequencing module is used for sequencing the vehicle track data according to time;
the first judgment module is used for regarding each vehicle in the target vehicle set, and if the speeds of two adjacent track points are all zero and the time difference between the two adjacent track points is within a preset time range, taking the previous track point in the two adjacent track points as a starting stop point;
the second judgment module is used for judging whether the vehicle stops at the interest point or whether the distance between the vehicle and the interest point is within a second preset distance or not if the subsequent stop point of the current vehicle is within a first preset distance from the start stop point and the duration time within the first preset distance exceeds a first preset duration;
the first determining module is used for determining the vehicle track data corresponding to the current vehicle as an original parking sample corresponding to the vehicle if the current vehicle does not park at the interest point or the distance between the current vehicle and the interest point is greater than the second preset distance.
In one or more embodiments, the second obtaining unit 704 includes:
the screening module is used for screening a track data sequence to be subjected to feature calculation with a second preset time length from the original docking sample;
the second determining module is used for classifying the track data sequence according to a target vehicle, vehicles around the target vehicle and a road where the target vehicle is located to obtain a track data sequence corresponding to each topic type; wherein the target subject comprises a target vehicle, vehicles around the target vehicle and a road on which the target vehicle is located;
and the third determining module is used for determining the characteristic data of the track data sequence corresponding to each topic category.
In one or more embodiments, the third determining module includes:
the first processing subunit is used for converting whether the current vehicle has a collision alarm, whether the current vehicle is in a running state or a parking state, whether the current vehicle is in a double-flash state or not and whether the current target vehicle is in a single-flash state into binary-type first preprocessing data through a vehicle-mounted terminal of the target vehicle;
the second processing subunit is used for processing the current speed, the acceleration, the time difference between different tracks and the included angle between the vehicle head and the due north direction through an abnormal value, a box sealing process and a normalization process to obtain second preprocessing data;
and the first determining subunit is used for taking the first preprocessed data and the second preprocessed data as the feature data of the track data sequence corresponding to the subject category of the target vehicle.
In one or more embodiments, the third determining module further includes:
the second determining subunit is used for determining the current timestamp of the target vehicle, and taking the union of a preset time period before the time corresponding to the timestamp and a preset time period after the time corresponding to the timestamp as a screening time period;
the third determining subunit is used for determining a search area by taking the geographic position of the target vehicle as a circle center and a preset distance as a radius based on the screening time period;
and the fourth determining subunit is used for determining track points of the peripheral vehicle in the search area and determining the average speed of each track point as the average speed value of the peripheral vehicle.
And the fifth determining subunit is used for taking the average speed value as the characteristic data of the trajectory data sequence corresponding to the vehicle subject category around the target vehicle.
In one or more embodiments, the third determining module further includes:
and the third processing subunit is used for performing binarization processing on the road type of the target vehicle and whether an intersection exists around the target vehicle at the current point to obtain the characteristic data of the track data sequence corresponding to the road subject type of the target vehicle.
In one or more embodiments, the training unit 706 includes:
the first acquisition module is used for acquiring a sample track data set; wherein the sample trajectory data set comprises training samples and validation samples;
the configuration module is used for configuring a verification category label for whether a traffic accident occurs or not for each sample track data in the verification sample according to a traffic accident rule;
the training module is used for inputting training track data in the training sample and the corresponding accident judgment type label into an initialized gated neural network model for training to obtain a training output result, wherein in each training process of the gated neural network model, the accident judgment type label corresponding to the track data is determined according to the number of track point sequences and the number of characteristics corresponding to the training track data;
and the fourth determining module is used for determining to obtain the gated neural network model under the condition that the training output result indicates that a convergence condition is reached, wherein the convergence condition is used for indicating that the difference degree between the type label of the accident judgment of the determined training track data and the verification type label is smaller than or equal to a preset threshold value, and the verification type label is a type label which has the same track point sequence number and feature number with the training track data in the verification sample.
In one or more embodiments, the determining unit 708 includes:
the conversion module is used for converting the data to be predicted corresponding to the current vehicle into a feature matrix comprising the number of track point sequences and the number of features;
the first processing module is used for inputting the characteristic matrix serving as input characteristics into the first layer gated network module of the trained gated neural network model to obtain a first output tensor;
the second processing module is used for inputting the first output result to a second layer of gated network module which inputs the trained gated neural network model to obtain a first splicing tensor of a starting time and a second splicing tensor of an ending time;
the full-connection module is used for inputting the first splicing tensor and the second splicing tensor into a full-connection layer of the trained gated neural network model to obtain the probability of whether a current vehicle has a traffic accident or not;
and the fifth determining module is used for determining whether a traffic accident occurs according to the probability.
In one or more embodiments, the traffic accident prediction apparatus further includes:
and the learning unit is used for learning the super-parameters of the gated neural network model through a Bayesian parameter adjusting algorithm, and obtaining the trained gated neural network model when the target super-parameters are determined.
According to still another aspect of the embodiment of the present invention, there is also provided an electronic device for implementing the traffic accident prediction method, where the electronic device may be a terminal device or a server shown in fig. 8. The present embodiment takes the electronic device as an example for explanation. As shown in fig. 8, the electronic device comprises a memory 802 and a processor 804, the memory 802 having a computer program stored therein, the processor 804 being arranged to perform the steps of any of the above-described method embodiments by means of the computer program.
Optionally, in this embodiment, the electronic device may be located in at least one network device of a plurality of network devices of a computer network.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
the method comprises the steps of S1, obtaining vehicle track data corresponding to a target vehicle set in a preset time period, and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set;
s2, acquiring feature data corresponding to a target theme based on the original docking sample;
s3, training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model;
and S4, inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident.
Alternatively, it can be understood by those skilled in the art that the structure shown in fig. 8 is only an illustration, and the electronic device may also be a terminal device such as a smart phone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a palmtop computer, a Mobile Internet Device (MID), a PAD, and the like. Fig. 8 is a diagram illustrating a structure of the electronic device. For example, the electronics may also include more or fewer components (e.g., network interfaces, etc.) than shown in FIG. 8, or have a different configuration than shown in FIG. 8.
The memory 802 may be used to store software programs and modules, such as program instructions/modules corresponding to the traffic accident prediction method and apparatus in the embodiments of the present invention, and the processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 802, so as to implement the traffic accident prediction method. The memory 802 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 802 can further include memory located remotely from the processor 804, which can be connected to the terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof. The memory 802 may be used for storing the prediction result of the traffic accident, but is not limited thereto. As an example, as shown in fig. 8, the memory 802 may include, but is not limited to, a first obtaining unit 702, a second obtaining unit 704, a training unit 706, and a determining unit 708 of the traffic accident prediction apparatus. In addition, but not limited to, other module units in the traffic accident prediction apparatus may also be included, which is not described in detail in this example.
Optionally, the transmitting device 808 is used for receiving or sending data via a network. Examples of the network may include a wired network and a wireless network. In one example, the transmission device 808 includes a Network adapter (NIC) that can be connected to a router via a Network cable and other Network devices to communicate with the internet or a local area Network. In one example, the transmission device 808 is a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
In addition, the electronic device further includes: a display 808 for displaying the prediction result of the traffic accident; and a connection bus 810 for connecting the respective module parts in the above-described electronic apparatus.
In other embodiments, the terminal device or the server may be a node in a distributed system, where the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting a plurality of nodes through a network communication. The nodes may form a Peer-To-Peer (P2P) network, and any type of computing device, such as a server, a terminal, and other electronic devices, may become a node in the blockchain system by joining the Peer-To-Peer network.
According to an aspect of the application, a computer program product or computer program is provided, comprising computer instructions, the computer instructions being stored in a computer readable storage medium. A processor of the computer device reads the computer instructions from the computer-readable storage medium, the processor executing the computer instructions causing the computer device to perform the traffic accident prediction method described above, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when executed.
Alternatively, in the present embodiment, the above-mentioned computer-readable storage medium may be configured to store a computer program for executing the steps of:
s1, obtaining vehicle track data corresponding to a target vehicle set in a preset time period, and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set;
s2, acquiring feature data corresponding to a target theme based on the original docking sample;
s3, training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model;
and S4, inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident.
Alternatively, in this embodiment, a person skilled in the art may understand that all or part of the steps in the methods of the foregoing embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disks, read-Only memories (ROMs), random Access Memories (RAMs), magnetic or optical disks, and the like.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
The integrated unit in the above embodiments, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in the above computer-readable storage medium. Based on such understanding, the technical solution of the present invention may be substantially or partially implemented in the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, and including instructions for causing one or more computer devices (which may be personal computers, servers, or network devices) to execute all or part of the steps of the method according to the embodiments of the present invention.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely illustrative, and for example, a division of a unit is merely a division of a logic function, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that it is obvious to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and these modifications and improvements should also be considered as the protection scope of the present invention.
Claims (12)
1. A traffic accident prediction method, comprising:
acquiring vehicle track data corresponding to a target vehicle set in a preset time period, and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set;
acquiring feature data corresponding to a target subject based on the original docking sample;
training a preset gated neural network model according to the characteristic data corresponding to the target theme to obtain a trained traffic accident prediction model;
and inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident.
2. The method of claim 1, wherein the preprocessing the vehicle trajectory data to obtain raw parking samples corresponding to the target vehicle set comprises:
sequencing the vehicle track data according to time;
for each vehicle in the target vehicle set, if the speeds of two adjacent track points are all zero and the time difference between the two adjacent track points is within a preset time range, taking the previous track point in the two adjacent track points as a starting stop point;
if the subsequent stop point of the current vehicle is within a first preset distance from the starting stop point and the duration within the first preset distance exceeds a first preset duration, judging whether the vehicle stops at the interest point or whether the distance between the vehicle and the interest point is within a second preset distance;
and if the current vehicle does not stop at the interest point or the distance between the current vehicle and the interest point is greater than the second preset distance, determining the vehicle track data corresponding to the current vehicle as an original stop sample corresponding to the vehicle.
3. The method of claim 1, wherein obtaining feature data corresponding to a target subject based on the raw docking sample comprises:
screening out a track data sequence to be subjected to feature calculation with a second preset time length from the original docking sample;
classifying the track data sequence according to a target vehicle, vehicles around the target vehicle and a road where the target vehicle is located to obtain a track data sequence corresponding to each topic type; wherein the target subject comprises a target vehicle, vehicles around the target vehicle and a road on which the target vehicle is located;
and determining the characteristic data of the track data sequence corresponding to each topic category.
4. The method of claim 3, wherein determining feature data of the sequence of trajectory data corresponding to the subject category of the target vehicle comprises:
converting whether the current vehicle has a collision alarm, whether the current vehicle is in a running state or a parking state, whether the current vehicle is in a double-flash state or not and whether the current target vehicle is in a single-flash state into binary-type first preprocessing data through a vehicle-mounted terminal of the target vehicle;
processing the current speed and the acceleration of the target vehicle, the time difference between different tracks and the included angle between the vehicle head and the due north direction by an abnormal value, a box sealing process and a normalization process to obtain second preprocessing data;
and taking the first preprocessing data and the second preprocessing data as feature data of a track data sequence corresponding to the subject category of the target vehicle.
5. The method of claim 3, wherein determining feature data of a sequence of trajectory data corresponding to a vehicle subject category in the vicinity of the target vehicle further comprises:
determining a current timestamp of the target vehicle, and taking a union of a preset time period before a time corresponding to the timestamp and a preset time period after the time corresponding to the timestamp as a screening time period;
determining a search area by taking the geographic position of the target vehicle as a circle center and a preset distance as a radius based on the screening time period;
determining track points of the surrounding vehicles in the search area, and determining the average speed of each track point as the average speed value of the surrounding vehicles;
and taking the average speed value as the characteristic data of the track data sequence corresponding to the vehicle theme category around the target vehicle.
6. The method of claim 3, wherein determining feature data of a track data sequence corresponding to a road subject category in which the target vehicle is located further comprises:
and carrying out binarization processing on the type of the road where the target vehicle is located and whether an intersection exists around the target vehicle at the current point to obtain the characteristic data of the track data sequence corresponding to the subject type of the road where the target vehicle is located.
7. The method according to claim 1, wherein the training of the preset gated neural network model according to the feature data corresponding to the target topic to obtain the trained traffic accident prediction model comprises:
acquiring a sample track data set; wherein the sample trajectory data set comprises training samples and validation samples;
configuring a verification category label for whether a traffic accident happens or not for each sample track data in the verification sample according to a traffic accident rule;
inputting training track data in the training sample and the corresponding accident judgment type label into an initialized gated neural network model for training to obtain a training output result, wherein in each training process of the gated neural network model, the accident judgment type label corresponding to the track data is determined according to the number of track point sequences and the number of characteristics corresponding to the training track data;
and under the condition that the training output result indicates that the convergence condition is reached, determining to obtain the gated neural network model, wherein the convergence condition is used for indicating that the difference degree between the determined class label of the accident judgment of the training track data and the verification class label is smaller than or equal to a preset threshold value, and the verification class label is the class label which has the same track point sequence number and the same characteristic number with the training track data in the verification sample.
8. The method of claim 7, wherein the inputting data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model to determine whether the current vehicle has a traffic accident comprises:
converting data to be predicted corresponding to the current vehicle into a feature matrix comprising the number of track point sequences and the number of features;
inputting the feature matrix serving as input features into a first layer gating network module of the trained gating neural network model to obtain a first output tensor;
inputting the first output result to a second layer gating network module which inputs the trained gating neural network model to obtain a first splicing tensor of the starting time and a second splicing tensor of the ending time;
inputting the first splicing tensor and the second splicing tensor into a full connection layer of the trained gated neural network model to obtain the probability of whether a current vehicle has a traffic accident or not;
and determining whether a traffic accident occurs according to the probability.
9. The method of claim 7, further comprising:
and learning the super-parameters of the gated neural network model through a Bayesian parameter-adjusting algorithm, and obtaining the trained traffic accident prediction model when determining to obtain the target super-parameters.
10. A traffic accident prediction apparatus, comprising:
the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring vehicle track data corresponding to a target vehicle set in a preset time period and preprocessing the vehicle track data to obtain an original parking sample corresponding to the target vehicle set;
the second obtaining unit is used for obtaining feature data corresponding to a target theme based on the original parking sample;
the training unit is used for training a preset gated neural network model according to the characteristic data to obtain a trained traffic accident prediction model;
and the determining unit is used for inputting the data to be predicted corresponding to the current vehicle into the trained traffic accident prediction model so as to determine whether the current vehicle has a traffic accident.
11. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to execute the method of any of claims 1 to 9 by means of the computer program.
12. A computer-readable storage medium, comprising a stored program, wherein the program when executed performs the method of any of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211501436.XA CN115797403A (en) | 2022-11-28 | 2022-11-28 | Traffic accident prediction method and device, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211501436.XA CN115797403A (en) | 2022-11-28 | 2022-11-28 | Traffic accident prediction method and device, storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115797403A true CN115797403A (en) | 2023-03-14 |
Family
ID=85442198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211501436.XA Pending CN115797403A (en) | 2022-11-28 | 2022-11-28 | Traffic accident prediction method and device, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115797403A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115938128A (en) * | 2023-03-15 | 2023-04-07 | 天津所托瑞安汽车科技有限公司 | Traffic accident prediction method, device, terminal and storage medium |
| CN116343484A (en) * | 2023-05-12 | 2023-06-27 | 天津所托瑞安汽车科技有限公司 | Traffic accident identification method, terminal and storage medium |
-
2022
- 2022-11-28 CN CN202211501436.XA patent/CN115797403A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115938128A (en) * | 2023-03-15 | 2023-04-07 | 天津所托瑞安汽车科技有限公司 | Traffic accident prediction method, device, terminal and storage medium |
| CN115938128B (en) * | 2023-03-15 | 2023-10-03 | 天津所托瑞安汽车科技有限公司 | Traffic accident prediction method, device, terminal and storage medium |
| CN116343484A (en) * | 2023-05-12 | 2023-06-27 | 天津所托瑞安汽车科技有限公司 | Traffic accident identification method, terminal and storage medium |
| CN116343484B (en) * | 2023-05-12 | 2023-10-03 | 天津所托瑞安汽车科技有限公司 | Traffic accident identification method, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dong et al. | Characterizing driving styles with deep learning | |
| CN109754594B (en) | Road condition information acquisition method and equipment, storage medium and terminal thereof | |
| CN110329271B (en) | Multi-sensor vehicle driving detection system and method based on machine learning | |
| CN107784587B (en) | Driving behavior evaluation system | |
| Gu et al. | A novel lane-changing decision model for autonomous vehicles based on deep autoencoder network and XGBoost | |
| CN115797403A (en) | Traffic accident prediction method and device, storage medium and electronic device | |
| CN113155173B (en) | Perception performance evaluation method and device, electronic device and storage medium | |
| CN111325978B (en) | Whole-process monitoring and warning system and method for abnormal behaviors of vehicles on expressway | |
| US10783360B1 (en) | Apparatuses, systems and methods for vehicle operator gesture recognition and transmission of related gesture data | |
| CN112396093B (en) | Driving scene classification method, device and equipment and readable storage medium | |
| CN112447041B (en) | Method and device for identifying operation behavior of vehicle and computing equipment | |
| GB2573738A (en) | Driving monitoring | |
| CN110060484A (en) | A kind of real-time early warning system and method violating the regulations of the highway passenger transportation based on block chain | |
| CN111291697A (en) | Method and device for recognizing obstacle | |
| CN106340205A (en) | Traffic monitoring method and traffic monitoring apparatus | |
| Toghi et al. | A maneuver-based urban driving dataset and model for cooperative vehicle applications | |
| Bouhsissin et al. | Driver behavior classification: a systematic literature review | |
| CN112951271A (en) | Tunnel traffic safety warning method and system based on acoustic assistance | |
| CN108304852A (en) | The determination method, apparatus and storage medium of road section type, electronic device | |
| Zeng et al. | On the importance of contextual information for building reliable automated driver identification systems | |
| CN117087674A (en) | Early warning method and system based on road risk and driver behavior | |
| CN107539038A (en) | A kind of vehicle tyre pressure method for monitoring state and device | |
| CN106448161A (en) | Road monitoring method and road monitoring device | |
| Paredes et al. | Intelligent collision risk detection in medium-sized cities of developing countries, using naturalistic driving: A review | |
| Hou | Applications of big data technology in intelligent transportation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |