CN113806585B - Method and device for obtaining road section passing duration, electronic equipment and storage medium - Google Patents
Method and device for obtaining road section passing duration, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113806585B CN113806585B CN202111075689.0A CN202111075689A CN113806585B CN 113806585 B CN113806585 B CN 113806585B CN 202111075689 A CN202111075689 A CN 202111075689A CN 113806585 B CN113806585 B CN 113806585B
- Authority
- CN
- China
- Prior art keywords
- road
- attribute information
- target
- feature
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/686—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/687—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The disclosure provides a method, a device, electronic equipment and a storage medium for obtaining a road section passing duration, relates to the field of artificial intelligence, and further relates to the technical field of intelligent traffic, so as to at least solve the technical problem that the prediction accuracy of the road section passing duration is low in the existing scheme. The specific implementation scheme is as follows: acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section; the method comprises the steps of carrying out feature fusion on road attribute information, time attribute information and space attribute information by using a target neural network model, and determining the passing duration of a target road section corresponding to target time to be predicted, wherein the target neural network model is obtained by machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
Description
Technical Field
The disclosure relates to the field of artificial intelligence, and further relates to the technical field of intelligent traffic, in particular to a method, a device, electronic equipment and a storage medium for acquiring the traffic duration of a road section.
Background
When providing navigation service, in order to assist the user in planning the journey, accurate estimation of the travel route of the user is required, that is, accurate prediction of traffic conditions of each road section forming the route at different times in the future is required.
In the existing scheme, a traffic prediction model based on space-time attention is adopted, and the traffic capacity of the current road section in the future time is predicted by modeling the space-time relationship of the current road section. However, the existing scheme cannot be flexibly applied to different scenes, so that prediction accuracy of the road section passing duration is affected.
Disclosure of Invention
The disclosure provides a method, a device, electronic equipment and a storage medium for obtaining a road section passing duration, so as to at least solve the technical problem that the prediction accuracy of the road section passing duration is low in the existing scheme.
According to an aspect of the present disclosure, there is provided a method for obtaining a traffic duration of a road segment, including: acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section; the method comprises the steps of carrying out feature fusion on road attribute information, time attribute information and space attribute information by using a target neural network model, and determining the passing duration of a target road section corresponding to target time to be predicted, wherein the target neural network model is obtained by machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
According to still another aspect of the present disclosure, there is provided an apparatus for acquiring a traffic duration of a road segment, including: the acquisition module is used for acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of the target road section; the determining module is used for carrying out feature fusion on the road attribute information, the time attribute information and the space attribute information by adopting the target neural network model, and determining the passing duration of the target road section corresponding to the target moment to be predicted; the target neural network model is obtained through machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
According to still another aspect of the present disclosure, there is provided an electronic apparatus including: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method for obtaining the road segment traffic duration set forth in the present disclosure.
According to yet another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method for acquiring a road segment traffic duration set forth in the present disclosure.
According to yet another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, performs the method of obtaining a road segment traffic duration set forth in the present disclosure.
In the present disclosure, road attribute information, time attribute information, space attribute information and target time to be predicted of a target road segment are obtained; the road attribute information, the time attribute information and the space attribute information are subjected to feature fusion by adopting the target neural network model, so that the traffic duration of the target road section corresponding to the target moment to be predicted is determined, the purpose of accurately predicting the traffic duration of the target road section based on the feature fusion of the road attribute information, the time attribute information and the space attribute information by adopting the target neural network model is achieved, the effect of improving the prediction precision of the traffic duration of the target road section is achieved, and the technical problem that the prediction precision of the traffic duration of the road section is low in the existing scheme is solved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 is a schematic diagram of a method for obtaining a road segment traffic duration in a prior art scheme;
FIG. 2 is a block diagram of a hardware architecture of a computer terminal (or mobile device) for implementing a method of obtaining a road segment travel duration in accordance with an embodiment of the present disclosure;
FIG. 3 is a flow chart of a method of obtaining a road segment travel duration in accordance with an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of the structure of a threshold module according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a method of obtaining a road segment traffic duration according to an embodiment of the present disclosure;
fig. 6 is a block diagram of an apparatus for obtaining a road segment passage duration according to an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. Furthermore, 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.
In order to assist the user in planning the journey, when the travel route of the user needs to be estimated accurately, that is, the traffic condition of each road section forming the route needs to be predicted accurately at different times in the future. The road states of the same road section at different moments often have larger differences, for example, certain commute road sections are regularly congested in the morning and evening peak time of the working day, and are mostly in a smooth state in the morning and evening peak time of the weekend; and for example, road conditions different from traffic congestion at any other time period caused by some sudden event.
The existing scheme designs a traffic prediction model based on a space-time attention neural network, and predicts the future traffic capacity of a road section by modeling the space-time relationship of the road section. The spatial properties of road segments are mainly from modeling the upstream and downstream road relations of road segments in the road network or mining of neighboring road segments that have a significant impact on the current road segment. The time attributes of the road section mainly include: (1) modeling a road segment history contemporaneous traffic law; (2) Modeling the real-time traffic law of the current moment of the road segment.
Fig. 1 is a schematic diagram of a method for obtaining a traffic duration of a road section in a conventional scheme. As shown in fig. 1, in the conventional scheme, road attribute features, history rule features, real-time road status features and road upstream and downstream features are fused together, and the traffic duration of a road section is predicted by a multi-Layer permission (MLP). The historical rule features and the real-time road state features are time attribute information.
In the existing scheme, fusion of time attribute information cannot dynamically select the use history rule features or the real-time road state features for different scenes. For some traffic scenes with strong regularity, such as commute scenes or scenes with inaccurate historic rules caused by sudden traffic conditions, the time attribute information cannot be accurately identified and filtered, and therefore prediction results are inaccurate.
The existing method cannot be flexibly suitable for different traffic scenes, and has the technical problem that the prediction accuracy of the road section passing duration is low.
According to an embodiment of the present disclosure, a method for obtaining a road segment traffic duration is provided, and it should be noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different from that herein.
The method embodiments provided by the embodiments of the present disclosure may be performed in a mobile terminal, a computer terminal, or similar electronic device. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein. Fig. 2 shows a block diagram of a hardware architecture of a computer terminal (or mobile device) for implementing a method of acquiring a traffic duration of a road segment.
As shown in fig. 2, the computer terminal 200 includes a computing unit 201 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 202 or a computer program loaded from a storage unit 208 into a Random Access Memory (RAM) 203. In the RAM 203, various programs and data required for the operation of the computer terminal 200 can also be stored. The computing unit 201, ROM 202, and RAM 203 are connected to each other through a bus 204. An input/output (I/O) interface 205 is also connected to bus 204.
Various components in computer terminal 200 are connected to I/O interface 205, including: an input unit 206 such as a keyboard, a mouse, etc.; an output unit 207 such as various types of displays, speakers, and the like; a storage unit 208 such as a magnetic disk, an optical disk, or the like; and a communication unit 209 such as a network card, modem, wireless communication transceiver, etc. The communication unit 209 allows the computer terminal 200 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunication networks.
The computing unit 201 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of computing unit 201 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 201 performs the method of obtaining the road segment traffic duration described herein. For example, in some embodiments, the method of obtaining the road segment travel duration may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 208. In some embodiments, part or all of the computer program may be loaded and/or installed onto the computer terminal 200 via the ROM 202 and/or the communication unit 209. When the computer program is loaded into the RAM 203 and executed by the computing unit 201, one or more steps of the method of obtaining the road segment travel duration described herein may be performed. Alternatively, in other embodiments, the computing unit 201 may be configured to perform the method of obtaining the road segment travel duration in any other suitable way (e.g., by means of firmware).
Various implementations of the systems and techniques described here can be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
It should be noted here that, in some alternative embodiments, the electronic device shown in fig. 2 described above may include hardware elements (including circuits), software elements (including computer code stored on a computer readable medium), or a combination of both hardware and software elements. It should be noted that fig. 2 is only one example of a specific example, and is intended to illustrate the types of components that may be present in the above-described electronic device.
In the above-described operation environment, the present disclosure provides a method of acquiring a road segment passage duration as shown in fig. 3, which may be performed by a computer terminal or similar electronic device as shown in fig. 2. Fig. 3 is a flowchart of a method for obtaining a traffic duration of a road segment according to an embodiment of the present disclosure. As shown in fig. 3, the method may include the steps of:
step S30, obtaining road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section;
the target road section may be a road section of a to-be-predicted traffic duration. For example, when providing navigation service for the user, multiple navigation routes can be generated according to the starting position of the user and the end position marked by the user, and the target road section can be any road section in the navigation routes.
The above-described road attribute information may be used to indicate a road attribute of the target link, and for example, the road attribute information may include the number of lanes of the target link, the speed limit condition, the length, the traffic light condition, and the like. The road attribute information may be acquired from an electronic map.
The time attribute information can be used for determining the historical rule characteristics and the real-time road state characteristics of the target road section, and the time attribute information can be obtained from the user track data. For example, the historical regular characteristics of the target road segments may be characterized as regular congestion occurring at the early and late peaks of the weekday, and in a clear state during the early and late peak periods of the non-weekday. The real-time road state characteristics of the target link may be represented as a road state different from traffic jams of any other period due to an emergency.
The above-mentioned spatial attribute information may be obtained according to a spatial dependency relationship of the target road segment, where the spatial attribute information may be used to determine upstream and downstream characteristics of the road of the target road segment, where the upstream and downstream of the road of the target road segment are road segments spatially adjacent to the target road segment, and the unobstructed condition of the upstream and downstream road may affect prediction of the traffic duration of the target road segment.
The target time to be predicted is the time for predicting the traffic duration of the target road section. For example, the target time to be predicted may be the current travel time of the user, or may be the travel time of the user in a future period of time.
Step S32, carrying out feature fusion on road attribute information, time attribute information and space attribute information by adopting a target neural network model, and determining the passing duration of a target road section corresponding to the target moment to be predicted;
the target neural network model is obtained through machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
The predetermined geographic area may be any geographic area based on provincial division. For example, when the machine learning training is performed on the target neural network model by using the road network information of different provinces, effective user track data in the corresponding province area is screened, and then the multi-dimensional attribute information of different road sections and the actual passing time length corresponding to each road section are obtained. The training data set of the target neural network model is formed by utilizing the road network information of a plurality of provinces, so that the overfitting of the target neural network can be effectively prevented, and the prediction performance of the target neural network model is further improved.
The multi-dimensional attribute information includes road attribute information, time attribute information, and space attribute information.
The actual passage duration corresponding to each road section comprises a true passage duration value of the target road section in the user track data acquired from the electronic map.
Specifically, the implementation process of determining the traffic duration of the target road segment corresponding to the target moment to be predicted by performing feature fusion on the road attribute information, the time attribute information and the space attribute information by using the target neural network model may refer to further description of the embodiments of the present disclosure, and will not be repeated.
According to the above steps S30 to S32 of the present disclosure, road attribute information, time attribute information, space attribute information, and a target time to be predicted of a target road segment are obtained; the road attribute information, the time attribute information and the space attribute information are subjected to feature fusion by adopting the target neural network model, so that the traffic duration of the target road section corresponding to the target moment to be predicted is determined, the purpose of accurately predicting the traffic duration of the target road section based on the feature fusion of the road attribute information, the time attribute information and the space attribute information by adopting the target neural network model is achieved, the effect of improving the prediction precision of the traffic duration of the target road section is achieved, and the technical problem that the prediction precision of the traffic duration of the road section is low in the existing scheme is solved.
The method for obtaining the traffic duration of the road section according to the above embodiment is further described below.
As an optional implementation manner, the method for obtaining the road section passing duration further includes:
step S310, training an initial neural network model by utilizing multidimensional attribute information of different road sections in a preset geographic area to obtain predicted passing duration;
step S312, obtaining an error by comparing the predicted passing time length with the actual passing time length;
and step S314, adjusting the initial neural network model based on the error to obtain a target neural network model.
According to the steps S310 to S314, training the initial neural network model based on the multidimensional attribute information of different road segments in the preset geographic area to obtain the predicted passing duration, comparing the predicted passing duration with the actual passing duration to obtain an error, and adjusting the initial neural network model according to the error to obtain the target neural network model, so that the prediction accuracy of the target neural network model on the passing duration of the target road segment can be further improved.
As an optional implementation manner, in step S32, feature fusion is performed on the road attribute information, the time attribute information and the space attribute information by using the target neural network model, and determining the traffic duration of the target road segment includes:
Step S320, determining road attribute characteristics through road attribute information, determining historical rule characteristics and real-time road state characteristics through time attribute information, and determining road upstream and downstream characteristics through space attribute information;
optionally, user track data acquired from the electronic map, traffic capacity information of different road segments in the past several months is extracted based on the user track data, and historical rule features of the target road segments are determined from the time attribute information through the first deep neural network model.
Optionally, user track data obtained from the electronic map, traffic capacity information of different road segments at the current moment and in a period of time nearby is extracted based on the user track data, and real-time road state characteristics are determined from the time attribute information through the second deep neural network model.
It should be noted that, the implementation process of determining the historical rule characteristics of the target road section by using the first deep neural network model and determining the real-time road state characteristics of the target road section by using the second neural network model may refer to the characteristic determining process of the existing scheme, which is not repeated.
Step S322, determining the feature to be used by utilizing the feature fusion result of the first original input feature and the second original input feature;
The first original input feature is an original input feature corresponding to a history rule feature, the second original input feature is an original input feature corresponding to a real-time road state feature, and the feature to be used is any one of the history rule feature and the real-time road state feature.
Optionally, the history rule feature may be obtained by processing the first original input feature through the first feature extraction network, and the real-time road state feature may be obtained by processing the second original input feature through the second feature extraction network.
It should be noted that, the specific processing procedures of the first feature extraction network and the second feature extraction network may refer to the feature processing procedures in the prior art, which are not described herein.
And step S324, adopting a target neural network model to perform feature fusion on the road attribute features, the road upstream and downstream features and the features to be used, and determining the passing duration of the target road section.
Through the steps S320 to S324 of the present disclosure, the characteristic fusion is performed on the road attribute characteristic, the road upstream and downstream characteristic and the to-be-used characteristic by using the target neural network model, the passing duration of the target road section is determined, and the to-be-used characteristic is flexibly adjusted, so that the dynamic adjustment of the characteristic fusion parameters according to different actual road conditions is realized, and the target neural network model has the capability of dynamically predicting the passing duration of the target road section.
As an alternative embodiment, in step S322, determining the feature to be used using the feature fusion result of the first original input feature and the second original input feature includes:
step S3220, determining a target threshold value by using the feature fusion result of the first original input feature and the second original input feature;
alternatively, the target threshold may be obtained according to the output result of the threshold module. The threshold module can identify the current scene by capturing the historical rule characteristics and the real-time road state characteristics of the target road section, and fusion of time attribute information is carried out.
The target threshold may be used to determine the tendency of the current target neural network model to historical rule features and real-time road state features, respectively. The target threshold value can enable the target neural network model to have different tendencies to the historical rule characteristics and the real-time road state characteristics under different scenes, and the prediction accuracy of the target neural network is further improved.
For example, the target threshold includes a real-time road state feature threshold of 0.9 and a history rule feature threshold of 0.1 if the real-time road state feature is preferred; and vice versa. When calculating the target threshold value, the sum of the real-time road state characteristic threshold value and the historical rule characteristic threshold value can be not 1, or can be a non-independent value, and can participate in calculation in the form of a hidden vector.
Step S3222, selecting the feature to be used from the history rule feature and the real-time road state feature based on the target threshold.
As an alternative embodiment, determining the target threshold using the feature fusion result of the first original input feature and the second original input feature includes:
step S40, mapping the first original input feature into a first hidden vector, mapping the second original input feature into a second hidden vector, and mapping the prediction time window information into a third hidden vector, wherein the prediction time window information is determined according to the target moment to be predicted;
the prediction time window information may indicate a sequence number of the prediction time window. For example, a window length of 5 minutes is a prediction time window, and when predicting the next 5 minutes, the prediction time window information is 1; when 60 minutes into the future is predicted, the predicted time window information is 12. The target neural network model may output the duration of the pass within each predicted time window.
Optionally, the first original input features are mapped to first hidden vectors and the second original input features are mapped to second hidden vectors using a full link layer (Fully Connected Layers, FC). Wherein the full link layer can be used to map the input features into hidden vectors in a high-dimensional space, i.e., to convert the explicit features into implicit features that can be learned by the target neural network model, which can alternatively be described as hidden vectors.
Optionally, the prediction time window information is mapped to a third stealth amount using an embedding layer (embedding). For example, every 5 minutes is a prediction time window, so that each prediction time window corresponds to a hidden vector. The third hidden vector may be continuously optimized in the training of the target neural network model.
Step S42, vector splicing processing is carried out on the first hidden vector, the second hidden vector and the third hidden vector, and a splicing result is obtained;
and S44, performing feature fusion processing on the splicing result to obtain a target threshold value.
As an alternative embodiment, the representation of the target threshold includes one of: explicit numerical representation, implicit vector representation.
Wherein the interpretability of the target neural network model can be enhanced in an explicit numerical representation.
For example, the explicit numerical expression is that all hidden vectors are multiplied by 0.1, and when the tendency of the history rule feature is stronger, the second hidden vector corresponding to the history rule feature is multiplied by 0.9, or by 1.2.
The implementation of determining the target threshold using the feature fusion result of the first original input feature and the second original input feature is described below in conjunction with fig. 4.
Fig. 4 is a schematic structural diagram of a threshold module according to an embodiment of the disclosure, as shown in fig. 4, in the threshold module, a first original input feature is mapped to a first hidden vector through a full link layer, a second original input feature is mapped to a second hidden vector through a full link layer, and prediction time window information is mapped to a third hidden vector. Further, vector splicing processing is carried out on the first hidden vector, the second hidden vector and the third hidden vector to obtain a spliced result, and feature fusion processing is carried out on the spliced result by using the multi-layer perceptron MLP to obtain a target threshold delta (.). Wherein, the multi-layer perceptron is a series connection of multi-layer full-link networks, the target threshold delta (·) can be expressed as a sigmoid function, the value range is (0, 1), and the target threshold delta (·) can be expressed by the following formula (1):
delta (·) =sigmoid (·) formula (1)
The following describes a process of implementing the feature fusion of road attribute information, time attribute information and space attribute information by using the target neural network model to determine the traffic duration of the target road segment corresponding to the target time to be predicted with reference to fig. 5.
Fig. 5 is a schematic diagram of a method for obtaining a road segment traffic duration according to an embodiment of the disclosure. As shown in fig. 5, the threshold module determines the current scene by using the obtained first original input feature and the second original input feature, and determines the tendency of the current target neural network model to the history rule feature and the real-time road state feature, respectively, where the tendency is expressed by the target threshold. The target threshold value output by the threshold module acts on the historical rule feature and the road real-time state feature respectively, so that the historical traffic rule of more dependent roads is dynamically selected when the future traffic time of the roads is calculated, or the road real-time state is realized, namely, the feature to be used is determined by utilizing the feature fusion result of the first original input feature and the second original input feature. And carrying out feature fusion on the road attribute features, the road upstream and downstream features and the features to be used by adopting a target neural network model, and determining the passing duration of the target road section.
The method and the device identify the road state of the road section at the current moment by modeling the historical rule of the road and the real-time road state, and dynamically perform feature fusion on the historical rule features and the real-time road state features. By dynamically using the historical rule information and the real-time road state information, the accuracy of traffic prediction of the target neural network model is improved, and the prediction capability of the target neural network model for handling emergencies is improved.
After the method for obtaining the road section passing duration provided by the present disclosure is applied, the accuracy of the estimated arrival time (Estimated Time Of Arrival, ETA) is improved, and the estimated effect under the scene with a large difference between the history passing rule and the real-time passing capability is obviously improved. The ETA perception and the use experience of the user can be obviously improved after the estimation of the route is accurate, better route planning service is provided for the user, the rationality of the user in road selection is ensured, and the travel of the user is scientifically guided.
The method comprises the steps of obtaining road attribute information, time attribute information, space attribute information of a target road section and target moment to be predicted; the road attribute information, the time attribute information and the space attribute information are subjected to feature fusion by adopting the target neural network model, so that the traffic duration of the target road section corresponding to the target moment to be predicted is determined, the purpose of accurately predicting the traffic duration of the target road section based on the feature fusion of the road attribute information, the time attribute information and the space attribute information by adopting the target neural network model is achieved, the effect of improving the prediction precision of the traffic duration of the target road section is achieved, and the technical problem that the prediction precision of the traffic duration of the road section is low in the existing scheme is solved.
In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.
From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present disclosure may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method described in the various embodiments of the present disclosure.
The disclosure further provides a device for obtaining the traffic duration of the road section, which is used for implementing the foregoing embodiments and the preferred embodiments, and the description thereof is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
Fig. 6 is a block diagram of an apparatus for acquiring a traffic duration of a road segment according to an embodiment of the present disclosure, and as shown in fig. 6, an apparatus 600 for acquiring a traffic duration of a road segment includes: the acquisition module 601 and the determination module 602.
The acquiring module 601 is configured to acquire road attribute information, time attribute information, space attribute information, and a target moment to be predicted of a target road segment;
the determining module 602 is configured to perform feature fusion on the road attribute information, the time attribute information and the space attribute information by using a target neural network model, and determine a traffic duration of a target road segment corresponding to a target moment to be predicted; the target neural network model is obtained through machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
Optionally, the apparatus 600 for obtaining the road segment traffic duration further includes: the training module 603 is configured to train the initial neural network model by using multidimensional attribute information of different road segments in a preset geographic area, so as to obtain a predicted passing duration; a comparison module 604, configured to obtain an error by comparing the predicted traffic duration with the actual traffic duration; and the adjustment module 605 is configured to adjust the initial neural network model based on the error, so as to obtain a target neural network model.
Optionally, the determining module 602 is configured to perform feature fusion on the road attribute information, the time attribute information, and the space attribute information by using the target neural network model, and determining the traffic duration of the target road segment includes: determining road attribute characteristics through road attribute information, determining historical rule characteristics and real-time road state characteristics through time attribute information, and determining road upstream and downstream characteristics through space attribute information; determining a feature to be used by utilizing a feature fusion result of a first original input feature and a second original input feature, wherein the first original input feature is an original input feature corresponding to a history rule feature, the second original input feature is an original input feature corresponding to a real-time road state feature, and the feature to be used is any one of the history rule feature and the real-time road state feature; and carrying out feature fusion on the road attribute features, the road upstream and downstream features and the features to be used by adopting a target neural network model, and determining the passing duration of the target road section.
Optionally, the determining module 602 is further configured to determine the feature to be used by using a feature fusion result of the first original input feature and the second original input feature includes: determining a target threshold value by utilizing a feature fusion result of the first original input feature and the second original input feature; and selecting the feature to be used from the historical rule feature and the real-time road state feature based on the target threshold.
Optionally, the determining module 602 further is configured to determine the target threshold using a feature fusion result of the first original input feature and the second original input feature includes: mapping the first original input features into first hidden vectors, mapping the second original input features into second hidden vectors, and mapping prediction time window information into third hidden vectors, wherein the prediction time window information is determined according to target time to be predicted; vector splicing processing is carried out on the first hidden vector, the second hidden vector and the third hidden vector, and a splicing result is obtained; and carrying out feature fusion processing on the splicing result to obtain a target threshold value.
Optionally, the representation of the target threshold includes one of: explicit numerical representation, implicit vector representation.
It should be noted that each of the above modules may be implemented by software or hardware, and for the latter, it may be implemented by, but not limited to: the modules are all located in the same processor; alternatively, the above modules may be located in different processors in any combination.
According to an embodiment of the present disclosure, there is also provided an electronic device comprising a memory having stored therein computer instructions and at least one processor configured to execute the computer instructions to perform the steps of any of the method embodiments described above.
Optionally, the electronic device may further include a transmission device and an input/output device, where the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in the present disclosure, the above processor may be configured to perform the following steps by a computer program:
s1, acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section;
and S2, carrying out feature fusion on the road attribute information, the time attribute information and the space attribute information by adopting a target neural network model, and determining the passing duration of the target road section corresponding to the target moment to be predicted.
Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments and optional implementations, and this embodiment is not described herein.
According to an embodiment of the present disclosure, the present disclosure also provides a non-transitory computer readable storage medium having stored therein computer instructions, wherein the computer instructions are configured to perform the steps of any of the method embodiments described above when run.
Alternatively, in the present embodiment, the above-described nonvolatile storage medium may be configured to store a computer program for performing the steps of:
s1, acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section;
and S2, carrying out feature fusion on the road attribute information, the time attribute information and the space attribute information by adopting a target neural network model, and determining the passing duration of the target road section corresponding to the target moment to be predicted.
Alternatively, in the present embodiment, the non-transitory computer readable storage medium described above may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
According to an embodiment of the present disclosure, the present disclosure also provides a computer program product. Program code for carrying out the audio processing methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the foregoing embodiments of the present disclosure, the descriptions of the various embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present disclosure, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The 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 units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present disclosure may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present disclosure may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present disclosure. And the aforementioned storage medium includes: a usb disk, a read-only memory (ROM), a random-access memory (RAM), a removable hard disk, a magnetic disk, or an optical disk, etc., which can store program codes.
The foregoing is merely a preferred embodiment of the present disclosure, and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present disclosure, which are intended to be comprehended within the scope of the present disclosure.
Claims (7)
1. A method of obtaining a road segment traffic duration, comprising:
acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of a target road section;
determining historical rule characteristics and real-time road state characteristics through the time attribute information;
mapping a first original input feature into a first hidden vector, mapping a second original input feature into a second hidden vector, and mapping prediction time window information into a third hidden vector, wherein the first original input feature is an original input feature corresponding to the history rule feature, the second original input feature is an original input feature corresponding to the real-time road state feature, and the prediction time window information is determined according to the target moment to be predicted;
vector splicing processing is carried out on the first hidden vector, the second hidden vector and the third hidden vector to obtain a splicing result;
Performing feature fusion processing on the splicing result to obtain a target threshold;
selecting a feature to be used from the historical rule feature and the real-time road state feature based on the target threshold;
and carrying out feature fusion on the road attribute information, the space attribute information and the to-be-used features by adopting a target neural network model, and determining the passing duration of the target road section corresponding to the to-be-predicted target moment, wherein the target neural network model is obtained by machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
2. The method of claim 1, the method further comprising:
training an initial neural network model by utilizing the multidimensional attribute information of different road sections in the preset geographic area to obtain predicted passing duration;
obtaining an error by comparing the predicted passage duration with the actual passage duration;
and adjusting the initial neural network model based on the error to obtain the target neural network model.
3. The method of claim 1, wherein employing the target neural network model to perform feature fusion on the road attribute information, the time attribute information, and the space attribute information, determining a duration of passage of the target road segment comprises:
determining road attribute characteristics according to the road attribute information, and determining road upstream and downstream characteristics according to the space attribute information;
and carrying out feature fusion on the road attribute features, the road upstream and downstream features and the features to be used by adopting the target neural network model, and determining the passing duration of the target road section, wherein the features to be used are any one of the history rule features and the real-time road state features.
4. The method of claim 1, wherein the representation of the target threshold comprises one of: explicit numerical representation, implicit vector representation.
5. An apparatus for obtaining a road segment traffic duration, comprising:
the acquisition module is used for acquiring road attribute information, time attribute information, space attribute information and target moment to be predicted of the target road section;
the determining module is used for determining historical rule characteristics and real-time road state characteristics through the time attribute information;
The mapping module is used for mapping a first original input feature into a first hidden vector, mapping a second original input feature into a second hidden vector and mapping prediction time window information into a third hidden vector, wherein the first original input feature is an original input feature corresponding to the history rule feature, the second original input feature is an original input feature corresponding to the real-time road state feature, and the prediction time window information is determined according to the target moment to be predicted;
the splicing module is used for carrying out vector splicing processing on the first hidden vector, the second hidden vector and the third hidden vector to obtain a splicing result;
the fusion module is used for carrying out feature fusion processing on the splicing result to obtain a target threshold value;
the selection module is used for selecting a feature to be used from the historical rule feature and the real-time road state feature based on the target threshold;
the determining module is further used for carrying out feature fusion on the road attribute information, the space attribute information and the to-be-used features by adopting a target neural network model, and determining the passing duration of the target road section corresponding to the to-be-predicted target moment;
The target neural network model is obtained through machine learning training by using a plurality of groups of data, and each group of data in the plurality of groups of data comprises: the multi-dimensional attribute information of different road sections in the geographic area and the actual passing duration corresponding to each road section are preset.
6. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-4.
7. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111075689.0A CN113806585B (en) | 2021-09-14 | 2021-09-14 | Method and device for obtaining road section passing duration, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111075689.0A CN113806585B (en) | 2021-09-14 | 2021-09-14 | Method and device for obtaining road section passing duration, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113806585A CN113806585A (en) | 2021-12-17 |
| CN113806585B true CN113806585B (en) | 2023-10-13 |
Family
ID=78895321
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111075689.0A Active CN113806585B (en) | 2021-09-14 | 2021-09-14 | Method and device for obtaining road section passing duration, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113806585B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114550453B (en) * | 2022-02-23 | 2023-09-26 | 阿里巴巴（中国）有限公司 | Model training method, model determining method, electronic device and computer storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072240A1 (en) * | 2016-10-20 | 2018-04-26 | 中国科学院深圳先进技术研究院 | Direction-variable lane control method for tidal traffic flow on road network |
| CN111860763A (en) * | 2020-06-05 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Model training method, parameter prediction method, model training device, parameter prediction device, electronic equipment and storage medium |
| CN112634614A (en) * | 2020-12-16 | 2021-04-09 | 安徽百诚慧通科技有限公司 | Long downhill traffic incident real-time detection method, device and storage medium |
| CN112907958A (en) * | 2021-01-29 | 2021-06-04 | 北京百度网讯科技有限公司 | Road condition information determining method and device, electronic equipment and readable medium |
-
2021
- 2021-09-14 CN CN202111075689.0A patent/CN113806585B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072240A1 (en) * | 2016-10-20 | 2018-04-26 | 中国科学院深圳先进技术研究院 | Direction-variable lane control method for tidal traffic flow on road network |
| CN111860763A (en) * | 2020-06-05 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Model training method, parameter prediction method, model training device, parameter prediction device, electronic equipment and storage medium |
| CN112634614A (en) * | 2020-12-16 | 2021-04-09 | 安徽百诚慧通科技有限公司 | Long downhill traffic incident real-time detection method, device and storage medium |
| CN112907958A (en) * | 2021-01-29 | 2021-06-04 | 北京百度网讯科技有限公司 | Road condition information determining method and device, electronic equipment and readable medium |
Non-Patent Citations (3)
| Title |
|---|
| Lin Hu ; Yuanxing Zhong ; Wei Hao.Optimal Route Algorithm Considering Traffic Light and Energy Consumption.IEEE.2018,全文. * |
| 基于时空相关属性模型的公交到站时间预测算法;赖永炫;张璐;杨帆;卢卫;王田;;软件学报(第03期);全文 * |
| 基于神经网络的目标分群及目标运动状态预测;张永利;刘楠楠;王兆伟;;舰船电子对抗(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113806585A (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872535B (en) | Intelligent traffic passage prediction method, device and server | |
| CN110210604B (en) | Method and device for predicting movement track of terminal equipment | |
| CN111582559B (en) | Arrival time estimation method and device | |
| CN104346926B (en) | Running time Forecasting Methodology and device and related terminal device | |
| CN112435469A (en) | Vehicle early warning control method and device, computer readable medium and electronic equipment | |
| CN113643532B (en) | Regional traffic prediction method and device | |
| CN112712712A (en) | Method and device for determining vehicle queuing information, road side equipment and cloud control platform | |
| CN113806585B (en) | Method and device for obtaining road section passing duration, electronic equipment and storage medium | |
| CN113159403A (en) | Method and device for predicting pedestrian track at intersection | |
| CN114715145B (en) | Trajectory prediction method, device and equipment and automatic driving vehicle | |
| CN113899381B (en) | Method, apparatus, device, medium, and product for generating route information | |
| CN111862583B (en) | Traffic flow prediction method and device | |
| CN111613053B (en) | Traffic disturbance detection and analysis method, device, storage medium and terminal | |
| CN111667083B (en) | Method and device for determining estimated network taxi | |
| CN116308738B (en) | Model training method, business wind control method and device | |
| CN115206102B (en) | Method, device, electronic equipment and medium for determining traffic path | |
| CN113850297B (en) | Road data monitoring method and device, electronic equipment and storage medium | |
| CN116089875A (en) | Traffic flow prediction method, device and storage medium integrating multisource space-time data | |
| CN115440063A (en) | Traffic signal lamp control method and device, computer equipment and storage medium | |
| CN115278757A (en) | Method and device for detecting abnormal data and electronic equipment | |
| CN113723585A (en) | Population prediction method and device, electronic equipment and storage medium | |
| CN114067565B (en) | Method and device for determining congestion identification accuracy | |
| CN114970949B (en) | Method and device for predicting running speed, electronic device, and storage medium | |
| CN113851007B (en) | Time interval dividing method and device, electronic equipment and storage medium | |
| KR102484139B1 (en) | Method, apparatus and system for calculating insurance premiums for two-wheeled vehicles based on driving pattern information of two-wheeled vehicles using an artificial intelligence model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |