CN110956299A - Arrival time estimation method and device - Google Patents
Arrival time estimation method and device Download PDFInfo
- Publication number
- CN110956299A CN110956299A CN201811128947.5A CN201811128947A CN110956299A CN 110956299 A CN110956299 A CN 110956299A CN 201811128947 A CN201811128947 A CN 201811128947A CN 110956299 A CN110956299 A CN 110956299A
- Authority
- CN
- China
- Prior art keywords
- road section
- state information
- neighbor
- target road
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- 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
- G06N3/047—Probabilistic or stochastic networks
-
- 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
- G06N3/048—Activation functions
Abstract
The embodiment of the invention provides an arrival time estimation method and device, and relates to the technical field of intelligent traffic. The method comprises the steps of obtaining path information to be estimated, wherein the path information comprises a plurality of target road sections and first state information of each target road section; respectively determining neighbor road sections of all target road sections and second state information of all neighbor road sections; and constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into the estimation model, and obtaining the estimated arrival time corresponding to the path information. The device is used for executing the method. According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are determined, the input characteristics are constructed according to the first state information and the second state information of the target road sections, the arrival time of the path is estimated by using the estimation model, and when the arrival time is estimated, the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a method and a device for estimating arrival time.
Background
In the field of electronic maps and navigation, the time taken by a moving body (vehicle, pedestrian, etc.) from a starting point to an end point, given a route, is a very important technical indicator describing the associated costs and expenses of traveling. The estimate of this Time is called the Estimated Time of Arrival (ETA).
In a taxi taking scene, a route from a starting point to a terminal point is obtained through a route planning service. The route may be a route from the driver to the boarding point of the passenger, or a route from the driver to the destination of the passenger after the passenger is received. After the driving route is determined, the driving time of the driver on the driving route needs to be estimated. In the field, ETA can be modeled into a machine learning problem through a machine learning method, in an electronic map, a path is often divided into a plurality of links (target road sections) connected end to end, and arrival time is estimated through the physical attributes of each link by using a machine learning model. However, when the vehicle is actually driven, the target road segment connected to the planned route may also affect the vehicle driven on the planned route, thereby causing a problem of low accuracy when estimating the arrival time.
Disclosure of Invention
In view of the above, an objective of the embodiments of the present invention is to provide a method and an apparatus for estimating an arrival time, so as to solve the above technical problems.
In a first aspect, an embodiment of the present invention provides a method for estimating arrival time, including:
acquiring path information to be estimated, wherein the path information comprises a plurality of target road sections and first state information of each target road section;
respectively determining neighbor road sections of all target road sections and second state information of all neighbor road sections;
and constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into an estimation model, and obtaining estimated arrival time corresponding to the path information.
Further, the determining the neighboring road segments of the target road segments and the second state information of the neighboring road segments respectively includes:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
Further, the calculating the order information between each target road segment and the corresponding neighbor road segment respectively includes:
if the target road section and the corresponding neighbor road section are at least spaced by N road sections, the order information between the target road section and the corresponding neighbor road section is N +1, N is not less than 0, and N is an integer.
Further, the constructing an input feature according to the first state information and the second state information includes:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
Further, the calculating the correlation coefficient between each target road segment and each corresponding neighboring road segment respectively includes:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
Further, the calculating the corresponding probability distribution according to the correlation coefficient includes:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
Further, the method further comprises:
and constructing the pre-estimation model and a multilayer perceptron model corresponding to each order, and training the pre-estimation model and the multilayer perceptron model.
Further, the first status information includes any one or a combination of the following corresponding to the target road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section;
the second state information comprises any one or combination of the following corresponding to the neighbor road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section.
In a second aspect, an embodiment of the present invention provides an arrival time estimation apparatus, including:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring path information to be estimated, and the path information comprises a plurality of target road sections and first state information of each target road section;
the second acquisition module is used for respectively determining neighbor road sections of all target road sections and second state information corresponding to the neighbor road sections;
and the estimation module is used for constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into an estimation model and obtaining the estimated arrival time corresponding to the path information.
Further, the second obtaining module is specifically configured to:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
Further, the second obtaining module is specifically configured to:
if the target road section and the corresponding neighbor road section are at least spaced by N road sections, the order information between the target road section and the corresponding neighbor road section is N +1, N is not less than 0, and N is an integer.
Further, the estimation module is specifically configured to:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
Further, the estimation module is specifically configured to:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
Further, the estimation module is specifically configured to:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
Further, the apparatus further comprises:
and the training module is used for constructing the pre-estimation model and the multilayer perceptron model corresponding to each order and training the pre-estimation model and the multilayer perceptron model.
Further, the first status information includes any one or a combination of the following corresponding to the target road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section;
the second state information comprises any one or combination of the following corresponding to the neighbor road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section.
In a third aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor, the processor being capable of performing the method steps of the first aspect when invoked by the program instructions.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, including:
the non-transitory computer readable storage medium stores computer instructions that cause the computer to perform the method steps of the first aspect.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for estimating arrival time according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a topology of a router network according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of another arrival time estimation method according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an arrival time estimation apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Fig. 1 is a schematic flow chart of a method for estimating arrival time according to an embodiment of the present invention, as shown in fig. 1, the method includes:
step 101: the method comprises the steps of obtaining path information to be estimated, wherein the path information comprises a plurality of target road sections and first state information of each target road section.
In a specific implementation process, the estimation device may obtain path information of a path to be estimated, and since one path in the electronic map is often divided into a plurality of end-to-end road segments, the path information of the path to be estimated may include a plurality of target road segments constituting the path and first state information corresponding to each target road segment. It should be noted that the first state information may be any one or a combination of a width of the target road segment, a number of lanes, a grade, a number of traffic lights, and a real-time traffic speed, and in addition, the first state information may further include other information representing a state of the target road segment, which is not specifically limited in this embodiment of the present invention.
For example: for the map provider, after the user selects the departure point and the destination, the map provider can provide the user with a selectable route, after the user selects one route, the arrival time corresponding to the route can be estimated, and the time required by the user from the departure point to the destination is the arrival time. In the taxi taking software, when a user makes a taxi taking request, a driver for taking a taxi is specified for the user, and the time taken by the driver to reach the place where the passenger is located and the time taken by the passenger from the place of departure to the destination are calculated.
Step 102: and respectively determining the neighbor road sections of the target road sections and the second state information of the neighbor road sections.
In a specific implementation process, due to the complexity of a road network, most target road segments have corresponding neighbor road segments, fig. 2 is a schematic diagram of a road network topology provided by an embodiment of the present invention, as shown in fig. 2, there are A, B, C, D, E and F six road segments (links) in a road network, and assuming that we want to estimate the arrival time of the path a- > D- > E, a traditional machine learning model only extracts features of A, D, E three links. These characteristics include, but are not limited to, road width, number of lanes, grade, toll, real-time traffic speed, etc. However, if F is congested, the efficiency of the traffic leaving D is inevitably affected. Although both D and E are currently not congested, according to the case of F, D is likely to be congested in the near future, which is very important information for predicting the model. B. C, F may be considered to be the neighbor links of the target link, and although they are not on the traffic path, they will affect the link on the traffic path by passing through the topology of the road network. Therefore, it is necessary to acquire the second state information of the corresponding respective neighbor road segments of each target road segment. It should be noted that the parameter field specifically comprised by the second state information should coincide with the first state information.
Step 103: and constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into an estimation model, and obtaining estimated arrival time corresponding to the path information.
In a specific implementation process, an input characteristic is constructed through the obtained first state information and the second state information, the input characteristic can be a matrix, the input characteristic is input into the estimation model, the estimation model can carry out internal calculation according to the input characteristic, and an output result is obtained, and the output result is the estimated arrival time corresponding to the path information.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
On the basis of the above embodiment, the determining the neighboring road segments of each target road segment and the second state information of each neighboring road segment respectively includes:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
In a specific implementation process, each target road segment may correspond to a plurality of neighboring road segments, and generally, two links closer to each other have a greater degree of mutual influence. We consider two links that are directly connected to each other as a "1 st order neighbor relation," for example, link B and link C in fig. 2 are both neighbor segments of link a, and the order is 1. The specific calculation order information is that if link A can reach link B only through N other links in the shortest time, A and B are considered to be N +1 order neighbor relations, wherein N is not less than 0, and N is an integer. And presetting a preset order according to the actual situation, and reserving second state information of the neighbor road segments and the data road segments which correspond to the target road segments and are less than or equal to the preset order. Still taking fig. 2 as an example, assuming that the preset order is 2, for link a, link B and link C are 1-order neighbor segments of link a, and link F is 2-order neighbor segments of link a, so that second state information corresponding to link B, link C and link F can be retained. Since link D is the target link, link D will not be considered as a neighbor link for link a.
It should be noted that, when N is larger, the more links the model takes into account, the wider the coverage area, and the more information is obtained from the road network topology. At the same time, however, as N increases, the number of links that need to be computed is likely to increase exponentially, causing a dramatic increase in computational cost. In general, N may be set to a value between 3 and 10.
The embodiment of the invention calculates the order information between the adjacent road sections corresponding to each target road section, acquires the adjacent road sections with the preset order or less and the corresponding second state information, and controls the calculation complexity while ensuring the accuracy of the estimated arrival time.
On the basis of any of the above embodiments, the constructing an input feature according to the first state information and the second state information includes:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
In a specific implementation process, the first state information corresponding to the obtained multiple pieces of target information is matrixed to obtain a first feature vector, a horizontal vector of the first feature vector may be multiple parameter fields included in the first state information, and a column vector may represent each target link, which may be exchanged, but this is not limited in this embodiment of the present invention.
And calculating to obtain a correlation coefficient between each adjacent road section corresponding to each target road section according to the first state information corresponding to each target road section and the second state information between the corresponding adjacent road sections. The correlation coefficient is calculated to obtain probability distribution, it should be noted that each parameter field in the second state information has a probability distribution, and the sum of probability values corresponding to all neighboring road segments in each probability distribution is 1.
And taking the probability value as a weight, and performing weighted summation on the second state information of all the neighbor road sections corresponding to each target road section to obtain a second feature vector.
After the first feature vector and the second feature vector are obtained, the second feature vector is spliced behind the first feature vector to form the input feature, and therefore the input feature is also a vector matrix.
According to the embodiment of the invention, the input characteristics are formed by calculating the correlation coefficient, the probability distribution and the weighted sum between the target road section and the adjacent road section, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the adjacent path related to the path when the arrival time is estimated.
On the basis of any of the above embodiments, the calculating the correlation coefficient between each target segment and each corresponding neighbor segment respectively includes:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
In a specific implementation process, when the order information between each neighboring road segment and the corresponding target road segment is calculated, the second state information corresponding to the neighboring road segment and the first state information corresponding to the target road segment may be input into the multi-layer perceptron model corresponding to the order information, and the multi-layer perceptron module may perform calculation according to the input first state information and the input second state information, and output a correlation coefficient between the neighboring path and the corresponding target path. It should be noted that each step corresponds to a multilayer perceptron model, and the multilayer perceptron model is constructed in advance.
According to the embodiment of the invention, the correlation coefficient between the neighbor road section and the target road section is obtained through calculation of the multilayer perceptron model, so that the influence of the neighbor road section on the target road section can be objectively evaluated, and the accuracy of subsequent estimation of the arrival time is improved.
On the basis of any of the above embodiments, the calculating a corresponding probability distribution according to the correlation coefficient includes:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
In a specific implementation process, after the correlation coefficient between each neighbor road section and the corresponding target road section is obtained through calculation, the probability distribution between the neighbor road section and the target road section is obtained through calculation by utilizing a softmax function.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
Fig. 3 is a schematic flow chart of another arrival time estimation method according to an embodiment of the present invention, as shown in fig. 3, the method further includes:
step 100: and constructing the pre-estimation model and the multilayer perceptron model corresponding to each order, and training the pre-estimation model and the multilayer perceptron model.
In a specific implementation process, before the arrival time is estimated, an estimation model for estimating the arrival time and a multilayer perceptron model corresponding to each order need to be constructed, and the estimation model and the multilayer perceptron model are optimized and trained together.
It should be noted that the specific implementation of steps 101 to 103 in the embodiment of the present invention is consistent with the above embodiment, and is not described herein again.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
Fig. 4 is a schematic structural diagram of an arrival time estimation apparatus according to an embodiment of the present invention, as shown in fig. 4, the apparatus includes: a first acquisition module 401, a second acquisition module 402, and a prediction module 403, wherein,
the first obtaining module 401 is configured to obtain path information to be estimated, where the path information includes a plurality of target road segments and first state information of each target road segment; the second obtaining module 402 is configured to determine neighboring road segments of each target road segment and second state information corresponding to the neighboring road segments respectively; the estimation module 403 is configured to construct an input feature according to the first state information and the second state information, and input the input feature into an estimation model to obtain an estimated arrival time corresponding to the path information.
In a specific implementation process, the first obtaining module 401 may obtain path information of a path to be estimated, where a path in the electronic map is often divided into a plurality of end-to-end segments, and therefore, the path information of the path to be estimated may include a plurality of target segments constituting the path and first state information corresponding to each target segment. It should be noted that the first state information may be any one or a combination of a width of the target road segment, a number of lanes, a grade, a number of traffic lights, and a real-time traffic speed, and in addition, the first state information may further include other information representing a state of the target road segment, which is not specifically limited in this embodiment of the present invention. The second obtaining module 402 obtains second status information of each corresponding neighbor road segment of each target road segment. It should be noted that the parameter field specifically comprised by the second state information should coincide with the first state information. The estimation module 403 constructs an input feature through the obtained first state information and the second state information, where the input feature may be a matrix, inputs the input feature into the estimation model, and the estimation model may perform internal calculation according to the input feature and obtain an output result, where the output result is an estimated arrival time corresponding to the path information.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
On the basis of the foregoing embodiment, the second obtaining module is specifically configured to:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
In a specific implementation process, the second obtaining module respectively calculates order information between each target road section and the corresponding neighbor road section, sets a preset order in advance according to an actual situation, and retains second state information of the neighbor road section and the data road section which are corresponding to each target road section and are less than or equal to the preset order.
The embodiment of the invention calculates the order information between the adjacent road sections corresponding to each target road section, acquires the adjacent road sections with the preset order or less and the corresponding second state information, and controls the calculation complexity while ensuring the accuracy of the estimated arrival time.
On the basis of any of the above embodiments, the second obtaining module is specifically configured to:
if the target road section and the corresponding neighbor road section are at least spaced by N road sections, the order information between the target road section and the corresponding neighbor road section is N +1, N is not less than 0, and N is an integer.
On the basis of any of the above embodiments, the estimation module is specifically configured to:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
In a specific implementation process, the estimation module performs matrixing on first state information corresponding to the obtained target information to obtain a first feature vector. The pre-estimation module can calculate and obtain a correlation coefficient between each adjacent road section corresponding to each target road section according to the first state information corresponding to each target road section and the second state information between the corresponding adjacent road sections, and calculates the correlation coefficient to obtain probability distribution. And taking the probability value as weight, and performing weighted summation on the second state information of all the neighbor road sections corresponding to each target road section by the pre-estimation module to obtain a second feature vector. After the first feature vector and the second feature vector are obtained, the pre-estimation module splices the second feature vector behind the first feature vector to form input features, and therefore the input features are also a vector matrix.
According to the embodiment of the invention, the input characteristics are formed by calculating the correlation coefficient, the probability distribution and the weighted sum between the target road section and the adjacent road section, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the adjacent path related to the path when the arrival time is estimated.
On the basis of any of the above embodiments, the estimation module is specifically configured to:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
In a specific implementation process, when calculating the order information between each neighboring road segment and the corresponding target road segment, the estimation module may input the second state information corresponding to the neighboring road segment and the first state information corresponding to the target road segment into the multi-layer perceptron model corresponding to the order information, and the multi-layer perceptron module may calculate according to the input first state information and the input second state information and output a correlation coefficient between the neighboring path and the corresponding target path. It should be noted that each step corresponds to a multilayer perceptron model, and the multilayer perceptron model is constructed in advance.
According to the embodiment of the invention, the correlation coefficient between the neighbor road section and the target road section is obtained through calculation of the multilayer perceptron model, so that the influence of the neighbor road section on the target road section can be objectively evaluated, and the accuracy of subsequent estimation of the arrival time is improved.
On the basis of any of the above embodiments, the estimation module is specifically configured to:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
In a specific implementation process, after the correlation coefficient between each neighbor road section and the corresponding target road section is obtained through calculation, the estimation module calculates and obtains probability distribution between the neighbor road section and the target road section through the softmax function.
According to the embodiment of the invention, the neighbor road sections of all target road sections and the second state information of all the neighbor road sections are respectively determined, the input characteristics are constructed according to the first state information and the second state information of all the target road sections, the arrival time of the path is estimated by using the estimation model, and the estimation accuracy of the arrival time is higher due to the consideration of the state information of the neighbor path related to the path when the arrival time is estimated.
On the basis of any one of the above embodiments, the apparatus further includes:
and the training module is used for constructing the pre-estimation model and the multilayer perceptron model corresponding to each order, and training the pre-estimation model and the multilayer perceptron model.
On the basis of any one of the above embodiments, the first status information includes any one or a combination of the following corresponding to the target road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section;
the second state information comprises any one or combination of the following corresponding to the neighbor road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working process of the apparatus described above may refer to the corresponding process in the foregoing method, and will not be described in too much detail herein.
In summary, in the embodiments of the present invention, the neighboring road segments of each target road segment and the second state information of each neighboring road segment are determined respectively, the input feature is constructed according to the first state information and the second state information of each target road segment, and the arrival time of the path is estimated by using the estimation model.
Fig. 5 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 5, specifically including a processor 501, a memory 502, a bus 503 and a communication interface 504, where the processor 501, the communication interface 504 and the memory 502 are connected through the bus 503; the processor 501 is arranged to execute executable modules, such as computer programs, stored in the memory 502.
The Memory 502 may include a high-speed Random Access Memory (RAM) and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. The communication connection between the network element of the system and at least one other network element is realized through at least one communication interface 504 (which may be wired or wireless), and the internet, a wide area network, a local network, a metropolitan area network, and the like can be used.
The memory 502 is used for storing a program, and the processor 501 executes the program after receiving an execution instruction, and the method performed by the apparatus defined by the flow program disclosed in any of the foregoing embodiments of the present application may be applied to the processor 501, or implemented by the processor 501.
The processor 501 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 501. The Processor 501 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. 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 application 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 the memory 502, and the processor 501 reads the information in the memory 502 and completes the steps of the method in combination with the hardware.
The arrival time estimation method provided by the embodiment can be executed by the device.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (18)
1. A method for estimating time of arrival, comprising:
acquiring path information to be estimated, wherein the path information comprises a plurality of target road sections and first state information of each target road section;
respectively determining neighbor road sections of all target road sections and second state information of all neighbor road sections;
and constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into an estimation model, and obtaining estimated arrival time corresponding to the path information.
2. The method of claim 1, wherein the determining the neighboring segments of the target segments and the second status information of the neighboring segments respectively comprises:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
3. The method of claim 2, wherein separately calculating the rank information between each target segment and the corresponding neighbor segment comprises:
if the target road section and the corresponding neighbor road section are at least spaced by N road sections, the order information between the target road section and the corresponding neighbor road section is N +1, N is not less than 0, and N is an integer.
4. The method of claim 1, wherein constructing input features from the first state information and the second state information comprises:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
5. The method of claim 4, wherein calculating the correlation coefficient between each target segment and each corresponding neighbor segment comprises:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
6. The method of claim 4, wherein said calculating a corresponding probability distribution from said correlation coefficients comprises:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
7. The method of claim 1, further comprising:
and constructing the pre-estimation model and a multilayer perceptron model corresponding to each order, and training the pre-estimation model and the multilayer perceptron model.
8. The method according to claim 1, wherein the first status information comprises any one or a combination of the following corresponding to the target road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section;
the second state information comprises any one or combination of the following corresponding to the neighbor road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section.
9. A time-of-arrival estimation apparatus, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring path information to be estimated, and the path information comprises a plurality of target road sections and first state information of each target road section;
the second acquisition module is used for respectively determining neighbor road sections of all target road sections and second state information corresponding to the neighbor road sections;
and the estimation module is used for constructing input characteristics according to the first state information and the second state information, inputting the input characteristics into an estimation model and obtaining the estimated arrival time corresponding to the path information.
10. The apparatus of claim 9, wherein the second obtaining module is specifically configured to:
respectively calculating order information between each target road section and the corresponding neighbor road section;
and acquiring the neighbor road sections of which the order information is less than or equal to a preset order and corresponding second state information.
11. The apparatus of claim 10, wherein the second obtaining module is specifically configured to:
if the target road section and the corresponding neighbor road section are at least spaced by N road sections, the order information between the target road section and the corresponding neighbor road section is N +1, N is not less than 0, and N is an integer.
12. The apparatus of claim 9, wherein the estimation module is specifically configured to:
constructing a first feature vector according to the first state information corresponding to each target road section;
respectively calculating a correlation coefficient between each target road section and each corresponding neighbor road section;
calculating corresponding probability distribution according to the correlation coefficient;
carrying out weighted summation on second state information of each neighbor road section of the target road section according to the probability distribution to obtain a second feature vector;
and obtaining the input features according to the first feature vector and the second feature vector.
13. The apparatus of claim 12, wherein the estimation module is specifically configured to:
and acquiring a corresponding multilayer perceptron model according to the order information of each neighbor road section and the corresponding target road section, and calculating a correlation coefficient between the neighbor road section and the corresponding target road section by using the multilayer perceptron model.
14. The apparatus of claim 12, wherein the estimation module is specifically configured to:
and calculating corresponding probability distribution by utilizing a softmax function according to the correlation coefficient.
15. The apparatus of claim 9, further comprising:
and the training module is used for constructing the pre-estimation model and the sensor model corresponding to each order and training the pre-estimation model and the multilayer sensor model.
16. The apparatus according to claim 9, wherein the first status information includes any one or a combination of the following corresponding to the target segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section;
the second state information comprises any one or combination of the following corresponding to the neighbor road segment: the target road section width, the number of lanes, the grade, the number of traffic lights and the real-time traffic speed corresponding to each target road section.
17. An electronic device, comprising: a processor, a memory, and a bus, wherein,
the processor and the memory are communicated with each other through the bus;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any one of claims 1-8.
18. A non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811128947.5A CN110956299B (en) | 2018-09-26 | 2018-09-26 | Arrival time estimation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811128947.5A CN110956299B (en) | 2018-09-26 | 2018-09-26 | Arrival time estimation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110956299A true CN110956299A (en) | 2020-04-03 |
| CN110956299B CN110956299B (en) | 2022-08-30 |
Family
ID=69967850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811128947.5A Active CN110956299B (en) | 2018-09-26 | 2018-09-26 | Arrival time estimation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110956299B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113902427A (en) * | 2021-12-09 | 2022-01-07 | 腾讯科技（深圳）有限公司 | Method, device and equipment for determining estimated arrival time and storage medium |
| CN117271918A (en) * | 2023-11-06 | 2023-12-22 | 腾讯科技（深圳）有限公司 | Information processing method, device, equipment, medium and product |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140058652A1 (en) * | 2012-08-27 | 2014-02-27 | International Business Machines Corporation | Traffic information processing |
| CN105006147A (en) * | 2015-06-19 | 2015-10-28 | 武汉大学 | Road segment travel time deducing method based on road space-time incidence relation |
| CN107111794A (en) * | 2015-01-11 | 2017-08-29 | 微软技术许可有限责任公司 | Prediction and the changeability using the travel time in Map Services |
| CN108288096A (en) * | 2017-01-10 | 2018-07-17 | 北京嘀嘀无限科技发展有限公司 | Method and device for estimating journey time, model training |
-
2018
- 2018-09-26 CN CN201811128947.5A patent/CN110956299B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140058652A1 (en) * | 2012-08-27 | 2014-02-27 | International Business Machines Corporation | Traffic information processing |
| CN107111794A (en) * | 2015-01-11 | 2017-08-29 | 微软技术许可有限责任公司 | Prediction and the changeability using the travel time in Map Services |
| CN105006147A (en) * | 2015-06-19 | 2015-10-28 | 武汉大学 | Road segment travel time deducing method based on road space-time incidence relation |
| CN108288096A (en) * | 2017-01-10 | 2018-07-17 | 北京嘀嘀无限科技发展有限公司 | Method and device for estimating journey time, model training |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113902427A (en) * | 2021-12-09 | 2022-01-07 | 腾讯科技（深圳）有限公司 | Method, device and equipment for determining estimated arrival time and storage medium |
| CN117271918A (en) * | 2023-11-06 | 2023-12-22 | 腾讯科技（深圳）有限公司 | Information processing method, device, equipment, medium and product |
| CN117271918B (en) * | 2023-11-06 | 2024-03-08 | 腾讯科技（深圳）有限公司 | Information processing method, device, equipment, medium and product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110956299B (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110361024B (en) | Method and system for dynamic lane-level vehicle navigation with vehicle group identification | |
| US10809723B2 (en) | Method and apparatus for generating information | |
| JP5941987B2 (en) | Traffic prediction apparatus and method | |
| US9002633B2 (en) | Method of determining and validating navigational priority settings utilizing probe data | |
| CN111326015A (en) | Parking spot recommendation method and device | |
| CN110942211A (en) | Prediction arrival time prediction method and device based on deep neural network | |
| CN110956299B (en) | Arrival time estimation method and device | |
| CN108133484A (en) | Automatic Pilot processing method and processing device based on scene cut, computing device | |
| JP2015161557A5 (en) | ||
| CN111862583B (en) | Traffic flow prediction method and device | |
| CN113723652A (en) | Service data processing method and device, electronic equipment and storage medium | |
| CN111508225A (en) | Information processing method, traffic control method, information processing device, traffic control equipment and storage medium | |
| CN115164922A (en) | Path planning method, system, equipment and storage medium | |
| CN113223293B (en) | Road network simulation model construction method and device and electronic equipment | |
| CN112985442B (en) | Driving path matching method, readable storage medium and electronic device | |
| CN113108806A (en) | Path planning method, device, equipment and medium | |
| Lee et al. | Development of reinforcement learning-based traffic predictive route guidance algorithm under uncertain traffic environment | |
| CN110260864B (en) | Method and device for constructing optimal reference trail route and electronic equipment | |
| CN116046008A (en) | Situation awareness-based route planning method, system and efficiency evaluation device | |
| JP2000193470A (en) | Route searching device and method and medium storing program for route searching | |
| CN110986992A (en) | Navigation method and device for unmanned vending vehicle, electronic equipment and storage medium | |
| CN111121803B (en) | Method and device for acquiring common stop points of road | |
| CN110648546A (en) | Road condition early warning method and device | |
| KR101607384B1 (en) | Method and system for providing counterfactual travel times for alternative routes | |
| CN106781470B (en) | Method and device for processing running speed of urban road |
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 |