Search

CN-116453332-B - Determination method of road traffic state data and generation method of speed prediction model

CN116453332BCN 116453332 BCN116453332 BCN 116453332BCN-116453332-B

Abstract

The application discloses a method for determining road traffic state data and a method for generating a speed prediction model. The method comprises the steps of obtaining flow data and road attribute data of a road in a current time period, determining the road state type of the road in a target time period according to the flow data and the road attribute data, inputting the flow data and the road attribute data into a speed prediction model, outputting predicted speed data of the road in the target time period, and determining the road traffic state data of the road in the target time period according to the road state type and the predicted speed data. The application solves the technical problem that the road traffic state data is difficult to accurately determine due to the complex road attribute of the urban road.

Inventors

  • ZHAO JUN
  • YANG BAOCHUN
  • ZHANG LEI

Assignees

  • 阿里云计算有限公司

Dates

Publication Date
20260512
Application Date
20230317

Claims (14)

  1. 1. A method of determining road traffic state data, comprising: acquiring flow data and road attribute data of a road in a current time period; Determining a road state type of the road in a target time period according to the flow data and the road attribute data, wherein the time of day is divided into a plurality of time periods according to the nature of the day and the rule of road traffic states, and the target time period is a time period in which the current time period falls in the plurality of time periods; Inputting the flow data and the road attribute data into a speed prediction model, and outputting predicted speed data of the road in the target time period; and determining road traffic state data of the road in the target time period according to the road state type and the predicted speed data.
  2. 2. The method of claim 1, wherein determining the road status type of the road for a target period of time from the flow data and the road attribute data comprises: And determining a target leaf node in a decision tree model according to the road attribute data, the target time period to which the current time period belongs and the flow data, and determining the road state type indicated by the target leaf node as the road state type of the road in the target time period.
  3. 3. The method of claim 1, wherein the velocity prediction model is obtained by: Acquiring a preset speed prediction model, wherein the preset speed prediction model comprises a relation between predicted speed data and flow data; Fitting model parameters of the preset speed prediction model according to the historical speed data, the historical flow data and the historical road attribute data of the road in a target historical time period until a loss value of the preset speed prediction model meets a preset loss condition, wherein the target historical time period and the target time period are the same time period in different days, and the preset loss condition means that errors of the predicted speed data and the historical speed data of the road under different historical flow data are minimum; And taking the preset speed prediction model after parameter adjustment as the speed prediction model.
  4. 4. The method of claim 3, wherein the pre-set speed prediction model is a road traffic impedance function, and fitting model parameters of the pre-set speed prediction model according to historical speed data, historical flow data, and historical road attribute data of the road within a target historical period of time until a loss value of the pre-set speed prediction model meets a pre-set loss condition comprises: acquiring road length, free flow speed data and road traffic capacity data from the historical road attribute data, wherein the free flow speed data refers to vehicle running speed data under the condition that a road is free of congestion, and the road traffic capacity data refers to the maximum traffic flow borne by the road; Calculating the free running time of the road according to the road length of the road and the free flow speed data; Calculating the road traffic time of the road according to the road length of the road and the historical speed data; And fitting model parameters in the road traffic impedance function according to the road traffic time, the free running time, the road traffic capacity data and the historical flow data.
  5. 5. The method according to claim 3 or 4, wherein the speed prediction model is a speed model of a plurality of roads or a speed model of a single road, the speed prediction model is obtained by fitting model parameters of the preset speed prediction model according to historical speed data, historical flow data and historical road attribute data of the plurality of roads in the target historical period of time in the case that the speed prediction model is a speed model of a plurality of roads, and the speed prediction model is obtained by determining the preset speed prediction model according to data of a single road in the case that the speed prediction model is a speed model of a single road and fitting model parameters of the preset speed prediction model according to historical speed data, historical flow data and historical road attribute data of the single road in the historical period of time.
  6. 6. A method according to claim 3, wherein the historical traffic data for the road over the target historical period of time is obtained by: Determining flow data of the road at different historical moments according to geographic position service data of the vehicle on the road; Aggregating the flow data on the road at different historical moments according to the time periods to obtain initial flow data of a plurality of historical time periods; Expanding the initial flow data according to the historical flow data quantity of the bayonets on the road to obtain flow data after expanding samples in a plurality of historical time periods; And acquiring flow data in the same time period as the target time period from the flow data subjected to sample expansion in a plurality of historical time periods, and obtaining historical flow data of the road in the target historical time period.
  7. 7. A method according to claim 3, wherein the historical speed data of the road over the target historical time is obtained by: acquiring the speed data of the road at different historical moments and the confidence associated with the speed data; selecting speed data with confidence higher than preset confidence from the speed data at different historical moments; Aggregating the speed data higher than the preset confidence coefficient according to the time periods to obtain historical speed data of a plurality of historical time periods; and acquiring the speed data of the same time period as the target time period from the historical speed data of the plurality of historical time periods to obtain the historical speed data of the road in the target historical time.
  8. 8. The method of claim 1, wherein the traffic data for the current time period of the road is obtained by: Determining flow data of the road at different moments according to geographic position service data of the vehicle on the road; Aggregating the flow data of the roads at different moments according to the time periods to obtain initial flow data of a plurality of time periods; expanding the initial flow data according to the flow data quantity of the bayonets on the road to obtain flow data after expanding the sample in a plurality of time periods; and acquiring flow data of the current time period from the flow data subjected to sample expansion of a plurality of time periods.
  9. 9. A method of determining traffic state data, comprising: the cloud server receives traffic data and road attribute data of a road in a current time period; The cloud server determines the road state type and the predicted speed data of the road in a target time period according to the flow data and the road attribute data, determines the road traffic state data of the road in the target time period according to the road state type and the predicted speed data, marks the road traffic state data of the road in the target time period on a map, and obtains a marked map, wherein the time of day is divided into a plurality of time periods according to the nature of the day and the rule of the road traffic state, and the target time period is a time period in which the current time period falls in the plurality of time periods; And the cloud server returns the marked map to the client and displays the marked map through the client.
  10. 10. A method of generating a velocity prediction model, comprising: Obtaining a preset speed prediction model, wherein the preset speed prediction model comprises a relation between predicted speed data and flow data, the speed prediction model is used for outputting the predicted speed data of a road in a target time period according to the flow data and the road attribute data of the road in a current time period, the road attribute data is also used for determining the road state type of the road in the target time period, the road state type and the predicted speed data are jointly used for determining the road traffic state data of the road in the target time period, the time of day is divided into a plurality of time periods according to the nature of the day and the rule of the road traffic state, and the target time period is a time period falling into the current time period in the plurality of time periods; Fitting model parameters of the preset speed prediction model according to the historical speed data, the historical flow data and the historical road attribute data of the road in a target historical time period until a loss value of the preset speed prediction model meets a preset loss condition, wherein the preset loss condition means that errors of the predicted speed data and the historical speed data of the road under different historical flow data are minimum; And taking the preset speed prediction model after parameter adjustment as the speed prediction model.
  11. 11. The method of claim 10, wherein the pre-set speed prediction model is a road traffic impedance function, and fitting model parameters of the pre-set speed prediction model according to historical speed data, historical flow data, and historical road attribute data of the road within a target historical period of time until a loss value of the pre-set speed prediction model meets a pre-set loss condition comprises: acquiring road length, free flow speed data and road traffic capacity data from the historical road attribute data, wherein the free flow speed data refers to vehicle running speed data under the condition that a road is free of congestion, and the road traffic capacity data refers to the maximum traffic flow borne by the road; Calculating the free running time of the road according to the road length of the road and the free flow speed data; Calculating the road traffic time of the road according to the road length of the road and the historical speed data; And fitting model parameters in the road traffic impedance function according to the road traffic time, the free running time, the road traffic capacity data and the historical flow data.
  12. 12. A road traffic state data determining apparatus, characterized by comprising: The acquisition unit is used for acquiring the flow data and the road attribute data of the road in the current time period; a first determining unit, configured to determine a road state type of the road in a target time period according to the flow data and the road attribute data, where a time of day is divided into a plurality of time periods according to a property of the day and a rule of a road traffic state, and the target time period is a time period in which the current time period falls in the plurality of time periods; the input unit is used for inputting the flow data and the road attribute data into a speed prediction model and outputting predicted speed data of the road in the target time period; And the second determining unit is used for determining road traffic state data of the road in the target time period according to the road state type and the predicted speed data.
  13. 13. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform the method of determining road traffic state data according to any one of claims 1 to 9, or to perform the method of generating a speed prediction model according to claim 10 or 11.
  14. 14. An electronic device, comprising: A memory storing an executable program; A processor for executing the program, wherein the program executes the method of determining road traffic state data according to any one of claims 1 to 9 or the method of generating the speed prediction model according to claim 10 or 11 when executed.

Description

Determination method of road traffic state data and generation method of speed prediction model Technical Field The application relates to the technical field of road traffic, in particular to a method for determining road traffic state data and a method for generating a speed prediction model. Background In road traffic simulation, updating of road traffic time plays a key role in path selection, traffic distribution, traffic state evaluation and the like, and road traffic state data is used as data for determining the road traffic time, so that the accuracy of the road traffic time is determined. In the related art, there is a method for determining road traffic state data in an expressway scene, but in an urban road scene, road properties are more complex, and factors affecting traffic states are more diverse, such as real-time traffic capacity of a road, traffic light phase at a downstream intersection, vehicle type proportion, weather conditions, and the like. The road traffic state data of the urban road is difficult to accurately determine by adopting the road traffic state data determining method under the expressway scene, so that the road traffic time of the urban road is difficult to accurately predict. In view of the above problems, no effective solution has been proposed at present. Disclosure of Invention The embodiment of the application provides a method for determining road traffic state data and a method for generating a speed prediction model, which are used for at least solving the technical problem that the road traffic state data is difficult to accurately determine due to the fact that the road attribute of an urban road is complex. According to one aspect of the embodiment of the application, a method for determining road traffic state data is provided, which comprises the steps of obtaining flow data and road attribute data of a road in a current time period, determining the road state type of the road in a target time period according to the flow data and the road attribute data, wherein the target time period comprises the current time period, inputting the flow data and the road attribute data into a speed prediction model, outputting predicted speed data of the road in the target time period, and determining the road traffic state data of the road in the target time period according to the road state type and the predicted speed data. According to another aspect of the embodiment of the application, a method for determining road traffic state data is provided, which comprises the steps that a cloud server receives flow data and road attribute data of a road in a current time period, the cloud server determines the road state type and the predicted speed data of the road in a target time period according to the flow data and the road attribute data, determines the road traffic state data of the road in the target time period according to the road state type and the predicted speed data, marks the road traffic state data of the road in the target time period on a map to obtain a marked map, and returns the marked map to a client and displays the marked map through the client. According to another aspect of the embodiment of the application, a method for generating a speed prediction model is provided, which comprises the steps of obtaining a preset speed prediction model, wherein the preset speed prediction model comprises a relation between predicted speed data and flow data, fitting model parameters of the preset speed prediction model according to historical speed data, historical flow data and historical road attribute data of a road in a target historical time period until a loss value of the preset speed prediction model meets a preset loss condition, the preset loss condition means that errors of the predicted speed data and the historical speed data of the road under different historical flow data are minimum, and taking the preset speed prediction model with the adjusted parameters as the speed prediction model. According to another aspect of the embodiment of the application, a device for determining road traffic state data is provided, which comprises an acquisition unit for acquiring traffic data and road attribute data of a road in a current time period, a first determination unit for determining the road state type of the road in a target time period according to the traffic data and the road attribute data, wherein the target time period comprises the current time period, an input unit for inputting the traffic data and the road attribute data into a speed prediction model and outputting predicted speed data of the road in the target time period, and a second determination unit for determining the road traffic state data of the road in the target time period according to the road state type and the predicted speed data. According to another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium, the computer-r