CN111126311A

CN111126311A - Potential dangerous area identification method and device for urban road and electronic equipment

Info

Publication number: CN111126311A
Application number: CN201911367300.2A
Authority: CN
Inventors: 李旭
Original assignee: Zebra Network Technology Co Ltd
Current assignee: Zebra Network Technology Co Ltd
Priority date: 2019-12-26
Filing date: 2019-12-26
Publication date: 2020-05-08
Anticipated expiration: 2039-12-26
Also published as: CN111126311B

Abstract

The invention provides a method, a device, electronic equipment and a computer readable storage medium for identifying potential dangerous areas of an urban road, wherein the method for identifying the potential dangerous areas of the urban road comprises the following steps: acquiring real-time track data of a plurality of vehicles; calculating a driving behavior score of each vehicle based on the real-time trajectory data; whether the specific area belongs to the potential danger area at the specific time is judged based on the driving behavior scores of all vehicles in the specific area at the specific time. According to the method for identifying the potential dangerous area of the urban road, the potential dangerous area can be accurately identified, and advance and real-time forecast can be carried out.

Description

Potential dangerous area identification method and device for urban road and electronic equipment

Technical Field

The invention relates to the field of vehicles, in particular to a method and a device for identifying a potential dangerous area of an urban road, electronic equipment and a computer readable storage medium.

Background

The significant increase of the automobile holding capacity brings great challenges to the safety of the automobile, and the existing method for identifying the dangerous area generally combines the road section and the time point where the accident occurs, and identifies the dangerous area based on the frequency of the accident occurring at a specific time or adopts a clustering method of machine learning to form the center of a cluster as the dangerous area.

The existing method for identifying the dangerous area is completely based on historical data, the problem of insufficient data samples exists, and due to certain randomness and sporadic nature of the occurrence time and the occurrence place of the traffic accident, the traditional method cannot forecast the potential dangerous area in advance and cannot reflect the time-varying characteristics of the accident.

Disclosure of Invention

In view of the above, the present invention provides a method, an apparatus, an electronic device and a computer readable storage medium for identifying a potentially dangerous area of an urban road, which can accurately identify the potentially dangerous area and can perform advanced and real-time forecasting.

In order to solve the above technical problem, in one aspect, the present invention provides a method for identifying a potentially dangerous area of an urban road, including the following steps:

acquiring real-time track data of a plurality of vehicles;

calculating a driving behavior score of each vehicle based on the real-time trajectory data;

whether the specific area belongs to the potential danger area at the specific time is judged based on the driving behavior scores of all vehicles in the specific area at the specific time.

Further, the obtaining the driving behavior score of each vehicle based on the real-time trajectory data includes:

extracting a corresponding vectorized representation of the vehicle based on the real-time trajectory data of the vehicle;

for the vectorized representation, extracting a low-dimensional vectorized representation through a deep learning model;

and acquiring the driving behavior scores of the vehicle at each time stamp based on the low-dimensional vectorization representation.

Further, based on the real-time trajectory data, extracting its corresponding vectorized representation comprises the steps of:

dividing the real-time track data according to track points;

acquiring the driving states of the front and rear 3 adjacent track points;

the driving state of each timestamp is quantized into a vectorized representation.

Further, the driving state is any one of acceleration straight running, acceleration right turning, acceleration left turning, deceleration straight running, deceleration right turning, deceleration left turning, uniform speed straight running, uniform speed right turning, and uniform speed left turning.

Further, the quantizing the driving state of each timestamp into a quantized representation comprises:

forming a driving state sequence according to the driving states of every 3 adjacent track points based on the real-time track data;

for the driving state sequence, dividing the driving state sequence into a plurality of segments according to a time window;

for each segment, extracting a corresponding feature vector of the segment, wherein an element of the feature vector is one of the driving state mark quantities;

based on the real-time track data, carrying out driving state conversion;

quantizing the driving state of each timestamp into the vectorized representation according to the driving state transitions of the current and next timestamps recorded by each timestamp.

Further, based on the real-time trajectory data, performing a driving state transition includes:

determining the transition probability and the duration of the transition process between various driving states according to the historical track data of the vehicle and by combining the identity label and the time window label of the vehicle;

and performing the driving state transition according to the transition probability and the duration of the transition process based on the real-time trajectory data of the vehicle.

Further, the low-dimensional vectorized representation includes: the vectorized representation of the identity feature of the vehicle and the vectorized representation of the time feature of the vehicle.

Further, determining the weight W of each model layer of the deep learning model and the bias b of each model layer through a linear regression method, and extracting the low-dimensional vectorization representation based on the deep model.

Further, based on the historical driving overall scores of the vehicles, a linear regression coefficient omega of the deep learning model is determined through a linear regression method, and weighted summation is carried out based on the low-dimensional vectorization representation and the linear regression coefficient, so that the driving behavior scores of the vehicles at all time stamps are obtained.

In a second aspect, the present invention provides a potentially dangerous area identification apparatus for an urban road, comprising:

the system comprises an acquisition module, a storage module and a control module, wherein the acquisition module is used for acquiring real-time track data of a plurality of vehicles;

a calculation module for determining a driving behavior score for each vehicle based on the real-time trajectory data;

and the potential danger area identification module is used for judging whether the specific area at the specific time belongs to the potential danger area or not based on the driving behavior scores of all vehicles in the specific area at the specific time.

In a third aspect, the present invention provides an electronic device for identification of potentially dangerous areas of urban roads, comprising:

one or more processors;

one or more memories having computer readable code stored therein, which when executed by the one or more processors, causes the processors to perform the steps of:

acquiring real-time track data of a plurality of vehicles;

In a fourth aspect, the present invention provides a computer-readable storage medium comprising:

in which computer readable code is stored which, when executed by one or more processors, causes the processors to perform the steps of:

acquiring real-time track data of a plurality of vehicles;

The technical scheme of the invention at least has one of the following beneficial effects:

according to the method for identifying the potential dangerous area of the urban road, whether the specific area at the specific time belongs to the potential dangerous area or not is judged based on the driving behavior scores of all vehicles in the specific area at the specific time, the potential dangerous area can be accurately identified, and advance and real-time forecast can be carried out.

Drawings

FIG. 1 is a flow chart of a method for identifying potentially dangerous areas of an urban road according to an embodiment of the invention;

FIG. 2 is a flowchart of a method for identifying potentially dangerous areas of an urban road according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a gated recurrent neural unit (GRU) in accordance with an embodiment of the present invention;

FIG. 4 is a diagram illustrating the embodiment shown in FIG. 2;

FIG. 5 is a schematic diagram of a potentially dangerous area identification apparatus for urban roads according to an embodiment of the present invention;

fig. 6 is a schematic diagram of an electronic device for identifying a potentially dangerous area of an urban road according to an embodiment of the present invention.

Detailed Description

The following detailed description of embodiments of the present invention will be made with reference to the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

The method combines the track data generated in the driving process of the vehicle, realizes the overall evaluation of the driving behaviors of different road sections at different time by means of the driving behavior grading results of different vehicles, and forecasts the quality of the overall evaluation as a potential danger area.

The method constructs a deep learning model capable of extracting vehicle identity characteristics (distinguishing overall driving styles of different vehicles) and time characteristics (driving styles of the same vehicle at different times), and simultaneously realizes overall evaluation of vehicle driving behaviors of different road sections by combining extracted low-dimensional vectorization expression (low-dimensional characteristic vector), and simultaneously forecasts dangerous areas based on grading results of real-time driving behaviors.

First, a method for identifying a potentially dangerous area of an urban road according to an embodiment of the present invention will be described with reference to fig. 1.

As shown in fig. 1, the method for identifying a potentially dangerous area of an urban road according to the embodiment of the present invention includes:

step S1, real-time trajectory data of a plurality of vehicles is acquired.

Preferably, the real-time trajectory data of the vehicle is acquired while the denoising process is performed. Therefore, interference is avoided, and data are more accurate.

In step S2, the driving behavior score of each vehicle is calculated based on the real-time trajectory data.

According to some embodiments of the invention, a method of calculating a driving behavior score for each vehicle based on real-time trajectory data comprises:

step S21, extracting its corresponding vectorized representation based on the real-time trajectory data of the vehicle.

Optionally, step S21 specifically includes:

and step S211, dividing the real-time track data according to track points.

And step S212, acquiring the driving states of the front and rear 3 adjacent track points.

Optionally, the driving state may include any one of: accelerating straight running, accelerating right turning, accelerating left turning, decelerating straight running, decelerating right turning, decelerating left turning, uniform speed straight running, uniform speed right turning and uniform speed left turning.

For example, a driving state sequence for each trajectory (stroke) is set

Where ID is index of driving the vehicle, N is the number of driving state series, t_nThe timestamp is the nth timestamp (e.g., 1 minute timestamp, 1440 total days), S_nIs the driving state corresponding to the time stamp.

Step S213 quantizes the driving state of each timestamp into a vectorized representation.

Preferably, step S213 includes:

1) for the driving state sequence, dividing the driving state sequence into a plurality of segments according to a time window;

optionally, a driving state sequence is divided into a plurality of segments according to time windows

Each segment corresponds to a time window (△ T), for example, when the time window Δ T is 30min, I1440/30 is 48, i.e., 48 segments.

2) For each segment, its corresponding feature vector is extracted, the elements of which are one of the driving state flag quantities.

Optionally, for each segment, its corresponding feature vector v may be extracted_i，v_iThe element of (1) is one of 9 driving state (accelerating straight line, accelerating right turn, accelerating left turn, decelerating straight line, decelerating right turn, decelerating left turn, uniform straight line, uniform right turn, uniform left turn) flags.

3) And performing driving state conversion based on the real-time track data.

Optionally, the conversion method is:

a) and determining the transition probability and the duration of the transition process between various driving states according to the historical track data of the vehicle and by combining the identity label and the time window label of the vehicle.

For example, when the driving state is converted from the front track point to the rear track point, that is, when the 9 driving states are converted to the other 9 driving states, there are 81 conversion combinations, and the probability of transition between two driving states and the duration of the transition process can be statistically calculated according to the vehicle identity label and the time window label by combining all track (trip) data of the vehicle.

b) And based on the real-time track data of the vehicle, carrying out driving state conversion according to the conversion probability and the duration of the conversion process.

4) And quantizing the driving state of each time stamp into a vectorized representation according to the driving state conversion of the current time stamp and the next time stamp recorded by each time stamp.

Optionally, the driving state of each timestamp is quantized into a 81-dimensional vectorized representation xⁱ∈R^m(m＝81)。

In step S22, for the vectorized representation, a low-dimensional vectorized representation is extracted by a deep learning model.

The low-dimensional vectorized representation may include a vectorized representation of the identity feature of the vehicle and a time feature vectorized representation of the vehicle.

Preferably, a self-encoder model (auto-encoder deep learning model) is employed to extract the low-dimensional vectorized representation.

For example, the self-coder model, in the coding phase: x is the number of_iFor the original vector representation of the ith vehicle, y_i ¹,y_i ²,...,y_i ^oIs the vector output when the original characteristic vector of the corresponding vehicle is positioned at each

hidden layer

1,2ⁱ∈R^dAnd d < 81 in the dimension space, namely outputting the dimension space as shown in the formula (1).

Wherein W is the weight of each layer of the model, b is the bias of each layer, and sigma is the sigmoid activation function.

In the decoding stage of the self-encoder model, the input implicit feature vector is the output z of the encodingⁱDecoding the output to obtain a reconstructed feature vector

The corresponding output at the hidden layer o, o-1, o-2

Of course, the above is only an optional example, and the deep learning may also be an encoding-decoding model (encoder-decoder), an attention model (attention), i.e. any model that can extract a low-dimensional vectorization representation should be understood to be within the scope of the present invention.

Further, the auto-encoder model (auto-encoder) is processed by a GRU (gated recurrent neural unit), which is just an optional example, and a CNN (convolutional neural network), an RNN (recurrent neural network) BiRNN (bidirectional recurrent neural network)/LSTM (long short term memory network), or the like may also be used.

Further, the weight W of each model layer of the deep learning model and the bias b of each model layer are determined through a linear regression method, and the low-dimensional vectorization representation is extracted based on the deep model.

In step S23, the driving behavior score of each time stamp of the vehicle is acquired based on the low-dimensional vectorization representation.

Further, a linear regression coefficient omega of the deep learning model is determined through a linear regression method based on the historical driving overall scores of the vehicles, and weighted summation is carried out based on the low-dimensional vectorization representation and the linear regression coefficient, so that the driving behavior scores of the vehicles at all time stamps are obtained.

Optionally, the method for determining the linear regression coefficient ω of the deep learning model by a linear regression method is as follows: and forming a vector omega according to the weight coefficients of the characteristics, and scoring the whole driving of all the vehicles according to c, so that a linear regression coefficient omega can be obtained based on a linear regression equation Z omega-c.

The historical driving overall score can be obtained by evaluating through an existing evaluation method or provided by a third party, and can be evaluated by combining the historical safe driving distance, the accident frequency, the traffic violation frequency, the overspeed driving frequency and the like of the vehicle.

Therefore, the potential dangerous area can be accurately identified, and the condition that identification errors are caused by dangerous driving of part of vehicles is avoided.

In step S3, it is determined whether or not the specific area at a specific time belongs to a potentially dangerous area based on the driving behavior scores of all the vehicles in the specific area at the specific time.

Alternatively, if the driving behavior scores of all the vehicles in the specific area at the specific time are lower than the set value, it is determined that the specific area at the specific time belongs to the potential danger area, and if the driving behavior scores of all the vehicles in the specific area at the specific time are higher than the set value, it is determined that the specific area at the specific time belongs to the safe area.

Thereby, potential danger areas can be accurately identified and advanced and real-time forecasts can be made.

As an example, as shown in FIG. 2, there are two parts, an offline training part and an online forecasting part.

Wherein, the first part, off-line training part includes:

1) acquiring off-line track data of a vehicle;

for example, trajectory data of a sufficient number of networked vehicles is acquired.

2) Data preprocessing and vectorization representation;

the data preprocessing is to perform preliminary denoising preprocessing on the acquired data.

And vectorization expression is to divide the real-time track data according to track points, acquire the driving state of the front and rear 3 adjacent track points and quantize the driving state of each timestamp into one vectorization expression.

For example, the driving state of each timestamp is quantized to one 81-dimension (9 driving states (accelerated straight, accelerated right turn, accelerated left turn, accelerated right turn, and accelerated left turn),Deceleration straight line, deceleration right turn, deceleration left turn, uniform straight line, uniform right turn, uniform left turn) to another 9 state transitions)ⁱ∈R^m(m＝81)。

3) A deep learning model (extracting a low-dimensional vectorized representation by the deep learning model (self-encoder model));

the extraction of the low-dimensional vectorized representation from the encoder model is divided into 2 stages.

In the stage 1, a self-Encoder model (Auto-Encoder) of deep learning is utilized to extract low-dimensional vectorization expression, wherein W is the weight of each layer of the model, b is the bias of each layer, and sigma is a sigmoid activation function, and the details are as follows:

a) let x_iFor the original vector representation of the ith vehicle, y_i ¹,y_i ²,...,y_i ^oIs the vector output when the original characteristic vector of the corresponding vehicle is positioned at each

hidden layer

1,2ⁱ∈R^dAnd the dimension space d is less than 81, and the specific implementation process is as follows:

b) the implicit feature vector input in the decoding stage (Decode Step) is the coded output zⁱDecoding the output to obtain a reconstructed feature vector

The corresponding output at the hidden layer o, o-1, o-2

c) Combining the encoding and decoding results to construct an objective function, wherein u is the number of all vehicles and u is the number of all vehicles_iIs the ith vehicle, targetThe function is as follows (3):

stage 2, the features extracted from the encoder are connected to a gated recurrent neural unit (GRU), as shown in fig. 3 below.

Wherein τ is a time window, and the corresponding variable represents a vector representation of each hidden layer or output layer when the time is τ, and the specific definition is the same as the above description, and it can be known that the state of the next timestamp is related to the previous timestamp according to the principle of GRU, specifically as follows:

the final objective function to be optimized is further defined as follows:

wherein H_c(G^τ) Reflecting the overall driving style of different vehicles for the identity characteristic attributes of the vehicles, s^τ _i,jThe cosine similarity is adopted to measure the similarity of the driving characteristics of different vehicles in the same time window, and the method specifically comprises the following steps:

4) low-dimensional vectorized representation (including vehicle identity feature vectorized representation and vehicle time feature vectorized representation);

5) the low-dimensional vectorized representation is combined with the overall scoring of the combined vehicle using linear regression learning, low-dimensional feature vector (low-dimensional vectorized representation) weights and linear regression coefficients.

Linear regression finds the weights of the low-dimensional feature vectors (low-dimensional vectorization representation) and the linear regression coefficients as follows:

a) the objective function (8) is optimized by adopting a random gradient descent method, and when the deep learning model is converged, low-dimensional characteristic vectors (low-dimensional vectorization representation) z of different vehicles in different time windows can be obtained_i ^τAnd all the weight coefficients W and b.

b) The driving behavior characteristics of the ith vehicle throughout the day are collectively expressed as:

T_mthe behaviour characteristic z of all vehicles as a total time window_iForm matrix Z_N×d(N rows and d columns, N is the number of vehicles, d is the dimension of the low-dimensional feature vector), and is denoted as Z ═ Z₁ ^T；z₂ ^T；...；z_N ^T]Where the superscript T represents the transpose of the vector.

c) Forming a vector omega according to the weight coefficients of the characteristics_d×1，c_N×1And scoring the whole driving times of all vehicles, so that a linear regression coefficient omega can be obtained based on a linear regression equation Z omega-c.

And in the second part, online forecasting comprises the following steps:

1) acquiring real-time track data of different vehicles;

2) outputting parameters W and b based on a deep learning model and real-time track data of different vehicles, and performing data preprocessing and vectorization representation;

3) carrying out integral scoring on the whole road based on the linear regression coefficient omega and the low-dimensional vectorization representation;

when the vehicle-mounted system is deployed on line, all the weight parameters W and b are deployed at a vehicle end, the low-dimensional vectorization representation of the track data is extracted by combining with the real-time track data, the low-dimensional vectorization representation is subjected to weighted summation with the linear regression coefficient omega to obtain the driving behavior scores of different vehicle timestamps, the scoring result is transmitted back to a cloud service center of the vehicle-mounted machine, and the service center can judge the dangerous driving degree of a specific region at the moment according to the driving average score of the vehicle in the specific region at a specific time.

The driving behavior scores of different road sections at different times reflect the overall safe driving degree of the vehicle at the road section at the time, and the area with the low overall score is regarded as a potential danger area.

According to the above method, a potential danger area distribution diagram for a certain time period at a certain place is identified as shown in fig. 4.

4) And forecasting the potential dangerous area or road section in real time.

For example, real-time forecasting is performed by means of navigation map labeling, voice broadcasting and the like.

Next, referring to fig. 5, a potentially dangerous area identifying apparatus 1000 for an urban road according to an embodiment of the present invention will be described.

As shown in fig. 5, a potentially dangerous area identifying apparatus 1000 for an urban road according to an embodiment of the present invention includes:

an obtaining module 1001 configured to obtain real-time trajectory data of a plurality of vehicles;

a calculating module 1002, configured to determine a driving behavior score of each vehicle based on the real-time trajectory data;

a potential danger area identification module 1003, configured to determine whether a specific area at a specific time belongs to a potential danger area based on the driving behavior scores of all vehicles in the specific area at the specific time.

Further, the potentially dangerous area identifying apparatuses 1000 for urban roads may also be respectively used for corresponding steps in the potentially dangerous area identifying method for urban roads, and a detailed description thereof is omitted here.

Further, an electronic device for potential hazardous area identification of urban roads according to an embodiment of the present invention is described with reference to fig. 6.

As shown in fig. 6, the electronic device for identifying a potentially dangerous area of an urban road according to an embodiment of the present invention includes:

a processor 1401 and a memory 1402, in which memory 1402 computer program instructions are stored, wherein the computer program instructions, when executed by the processor, cause the processor 1401 to perform the steps of:

step S1, acquiring real-time track data of a plurality of vehicles;

step S2, calculating a driving behavior score of each vehicle based on the real-time trajectory data;

The various interfaces and devices described above may be interconnected by a bus architecture. A bus architecture may be any architecture that may include any number of interconnected buses and bridges. Various circuits of one or more Central Processing Units (CPUs), represented in particular by processor 1401, and one or more memories, represented by memory 1402, are coupled together. The bus architecture may also connect various other circuits such as peripherals, voltage regulators, power management circuits, and the like. It will be appreciated that a bus architecture is used to enable communications among the components. The bus architecture includes a power bus, a control bus, and a status signal bus, in addition to a data bus, all of which are well known in the art and therefore will not be described in detail herein.

The network interface 1403 may be connected to a network (e.g., the internet, a local area network, etc.), obtain relevant data from the network, and store the relevant data in the hard disk 1405.

The input device 1404 may receive various instructions from an operator and send them to the processor 1401 for execution. The input device 1404 may include a keyboard or a pointing device (e.g., a mouse, trackball, touch pad, or touch screen, among others.

The display device 1406 may display a result obtained by the processor 1401 executing the instruction.

The memory 1402 is used for storing programs and data necessary for operating the operating system, and data such as intermediate results in the calculation process of the processor 1401.

It will be appreciated that the memory 1402 in embodiments of the invention may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The nonvolatile memory may be a Read Only Memory (ROM), a Programmable Read Only Memory (PROM), an Erasable Programmable Read Only Memory (EPROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), or a flash memory. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. The memory 1402 of the apparatus and methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

In some embodiments, memory 1402 stores elements, executable modules or data structures, or a subset thereof, or an expanded set thereof as follows: an operating system 14021 and application programs 14014.

The operating system 14021 includes various system programs, such as a framework layer, a core library layer, a driver layer, and the like, for implementing various basic services and processing hardware-based tasks. The application 14014 includes various applications, such as a Browser (Browser), and the like, for implementing various application services. A program implementing a method according to an embodiment of the invention may be included in the application 14014.

When the processor 1401 calls and executes the application program and data stored in the memory 1402, specifically, the application program or the instruction stored in the application 14014, first, real-time trajectory data of a plurality of vehicles is acquired; then, calculating driving behavior scores of all vehicles based on the real-time track data; finally, whether the specific area belongs to the potential danger area at the specific time is judged based on the driving behavior scores of all vehicles in the specific area at the specific time.

The methods disclosed by the above-described embodiments of the present invention may be applied to the processor 1401, or may be implemented by the processor 1401. Processor 1401 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by instructions in the form of hardware integrated logic circuits or software in the processor 1401. The processor 1401 may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, and may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in a memory 1402, and a processor 1401 reads information in the memory 1402 and performs the steps of the above method in combination with hardware thereof.

It is to be understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or any combination thereof. For a hardware implementation, the processing units may be implemented within one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), general purpose processors, controllers, micro-controllers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof.

For a software implementation, the techniques described herein may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.

In addition, an embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the processor is caused to execute the following steps:

step S1, acquiring real-time track data of a plurality of vehicles;

Still further, the present invention also provides a program product comprising execution instructions stored in a readable storage medium. The at least one processor of the electronic device (which may be, for example, a server, a cloud server, or a part of a server, etc.) may read the execution instruction from the readable storage medium, and execute the execution instruction, so that the potential dangerous area identifying apparatus 1000 for an urban road implements the potential dangerous area identifying method for an urban road provided in the various embodiments described above.

In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, 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, devices or units, and may be in an electrical, mechanical or other form.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be physically included alone, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the transceiving method according to various embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for identifying a potentially dangerous area of an urban road, comprising:

acquiring real-time track data of a plurality of vehicles;

2. The method for identifying potentially dangerous areas of an urban road according to claim 1, wherein said obtaining a driving behavior score for each vehicle based on said real-time trajectory data comprises:

3. The method of identifying potential hazard zones as claimed in claim 2 wherein extracting their respective vectorized representations based on the real-time trajectory data comprises the steps of:

dividing the real-time track data according to track points;

acquiring the driving states of the front and rear 3 adjacent track points;

4. The method according to claim 3, wherein the driving state is any one of acceleration straight traveling, acceleration right turning, acceleration left turning, deceleration straight traveling, deceleration right turning, deceleration left turning, uniform speed straight traveling, uniform speed right turning, and uniform speed left turning.

5. The method of identifying potential hazard zones as claimed in claim 4 wherein said quantifying the driving state of each time stamp into a vectorized representation comprises:

based on the real-time track data, carrying out driving state conversion;

6. The method of claim 5, wherein performing a driving state transition based on the real-time trajectory data comprises:

7. The method of identifying potential hazard zones as claimed in claim 6, wherein said low-dimensional vectorized representation comprises: the vectorized representation of the identity feature of the vehicle and the vectorized representation of the time feature of the vehicle.

8. The method according to claim 7, wherein the weights W of each model layer of the deep learning model and the bias b of each model layer are determined by a linear regression method, and the low-dimensional vectorized representation is extracted based on the depth model.

9. The potential danger area identifying method according to claim 2, wherein a linear regression coefficient ω of the deep learning model is determined by a linear regression method based on a historical driving overall score of each vehicle, and a driving behavior score of each time stamp of the vehicle is obtained by weighted summation based on the low-dimensional vectorized representation and the linear regression coefficient.

10. An apparatus for identifying a potentially dangerous area of an urban road, comprising:

11. An electronic device for potential hazardous area identification of urban roads, comprising:

one or more processors;

acquiring real-time track data of a plurality of vehicles;

12. A computer readable storage medium having computer readable code stored therein, which when executed by one or more processors, causes the processors to perform the steps of:

acquiring real-time track data of a plurality of vehicles;