Method for obtaining road condition information, apparatus thereof, and storage medium

1. A method for generating road-condition information by a first server, the method comprising:

obtaining a number of vehicles passing a driving road segment within a duration;

determining whether the driving road segment is a sparse road segment by determining whether the number of vehicles is less than or equal to a vehicle threshold, and

in response to determining the number of vehicles is less than or equal to the vehicle threshold, determining the driving road segment is the sparse road segment and then selecting the driving road segment as a target road segment, and

in response to determining the number of vehicles is greater than the vehicle threshold, determining the driving road segment is not the sparse road segment and then disregarding the driving road segment;

obtaining first real-time driving data transmitted by a first vehicle passing the target road segment, and obtaining first driving characteristic-information of the target road segment based on the first real-time driving data;

obtaining second real-time driving data transmitted by a second vehicle passing a topology road segment, and obtaining second driving characteristic-information of the target road segment based on the second real-time driving data, the topology road segment being a road segment within a target range of the target road segment;

sending an attribute information search request to a second server separate from the first server, the second server including a road segment management server;

receiving, from the second server, attribute information of the topology road segment generated by the road segment management server, wherein the attribute information of the topology road segment generated by the road segment management server includes a length of the topology road segment or a road grade of the topology road segment; and

generating road-condition information of the target road segment by inputting at least the second driving characteristic-information of the second vehicle on the topology road segment and the attribute information of the topology road segment to a trained road-condition prediction model,

wherein the road-condition prediction model is configured to output a congestion state of the target road segment and is trained by:

obtaining first historical driving characteristic-information of a third vehicle passing the target road segment during a target period and second historical driving characteristic-information of a fourth vehicle passing the topology road segment during the target period;

obtaining a true road-condition probability value of the target road segment during the target period using the first historical driving characteristic-information and the second historical driving characteristic-information, including: matching the first historical driving characteristic-information and the second historical driving characteristic-information with an offline driving characteristic-information to obtain a matching result, and filtering non-driving information in the matching result to generate the true road-condition probability value; and

training the road-condition prediction model using a classification algorithm and based on the first historical driving characteristic-information, the second historical driving characteristic-information, and the true road-condition probability value,

wherein the road-condition prediction model comprises a congestion prediction model configured to output a first road-condition probability value and a clearness prediction model configured to output a second road-condition probability value, and the congestion state of the target road segment is determined based on the first road-condition probability value and the second road-condition probability value.

2. The method according to claim 1 , wherein: the method further comprises:

obtaining attribute information of the target road segment; and the obtaining road-condition information of the target road segment comprises:

obtaining the road-condition information of the target road segment based on the first driving characteristic-information, the second driving characteristic-information, the attribute information of the target road segment, and the attribute information of the topology road segment.

3. The method according to claim 2 , further comprising:

sending attribute information search requests of the target road segment and the topology road segment to the road segment management server.

4. A first server for obtaining road-condition information, comprising: a memory storing computer program instructions; and a processor coupled to the memory and, when executing the computer program instructions, configured to perform:

obtaining a number of vehicles passing a driving road segment within a duration;

determining whether the driving road segment is a sparse road segment by determining whether the number of vehicles is less than or equal to a vehicle threshold, and

in response to determining the number of vehicles is less than or equal to the vehicle threshold, determining the driving road segment is the sparse road segment and then selecting the driving road segment as a target road segment, and

in response to determining the number of vehicles is greater than the vehicle threshold, determining the driving road segment is not the sparse road segment and then disregarding the driving road segment;

obtaining first real-time driving data transmitted by a first vehicle passing the target road segment, and obtaining first driving characteristic-information of the target road segment based on the first real-time driving data;

obtaining second real-time driving data transmitted by a second vehicle passing a topology road segment, and obtaining second driving characteristic-information of the target road segment based on the second real-time driving data, the topology road segment being a road segment within a target range of the target road segment;

sending an attribute information search request to a second server separate from the first server, the second server including a road segment management server;

receiving, from the second server, attribute information of the topology road segment generated by the road segment management server wherein the attribute information of the topology road segment generated by the road segment management server includes a length of the topology road segment or a road grade of the topology road segment; and

generating road-condition information of the target road segment by inputting at least the second driving characteristic-information of the second vehicle on the topology road segment and the attribute information of the topology road segment to a trained road-condition prediction model,

wherein the road-condition prediction model is configured to output a congestion state of the target road segment and is trained by:

obtaining first historical driving characteristic-information of a third vehicle passing the target road segment during a target period and second historical driving characteristic-information of a fourth vehicle passing the topology road segment during the target period;

obtaining a true road-condition probability value of the target road segment during the target period using the first historical driving characteristic-information and the second historical driving characteristic-information, including: matching the first historical driving characteristic-information and the second historical driving characteristic-information with an offline driving characteristic-information to obtain a matching result, and filtering non-driving information in the matching result to generate the true road-condition probability value; and

training the road-condition prediction model using a classification algorithm and based on the first historical driving characteristic-information, the second historical driving characteristic-information, and the true road-condition probability value,

wherein the road-condition prediction model comprises a congestion prediction model configured to output a first road-condition probability value and a clearness prediction model configured to output a second road-condition probability value, and the congestion state of the target road segment is determined based on the first road-condition probability value and the second road-condition probability value.

5. The first server according to claim 4 , wherein: the processor is further configured to perform:

obtaining attribute information of the target road segment; and the obtaining road-condition information of the target road segment comprises:

obtaining the road-condition information of the target road segment based on the first driving characteristic-information, the second driving characteristic-information, the attribute information of the target road segment, and the attribute information of the topology road segment.

6. The first server according to claim 5 , wherein the processor is further configured to perform:

sending attribute information search requests of the target road segment and the topology road segment to the road segment management server.

7. A non-transitory computer-readable storage medium storing computer program instructions executable by at least one processor of a first server to perform:

obtaining a number of vehicles passing a driving road segment within a duration;

determining whether the driving road segment is a sparse road segment by determining whether the number of vehicles is less than or equal to a vehicle threshold, and

in response to determining the number of vehicles is less than or equal to the vehicle threshold, determining the driving road segment is the sparse road segment and then selecting the driving road segment as a target road segment, and

in response to determining the number of vehicles is greater than the vehicle threshold, determining the driving road segment is not the sparse road segment and then disregarding the driving road segment;

obtaining first real-time driving data transmitted by a first vehicle passing the target road segment, and obtaining first driving characteristic-information of the target road segment based on the first real-time driving data;

obtaining second real-time driving data transmitted by a second vehicle passing a topology road segment, and obtaining second driving characteristic-information of the target road segment based on the second real-time driving data, the topology road segment being an adjacent road segment within a target range of the target road segment;

sending an attribute information search request to a second server separate from the first server, the second server including a road segment management server;

receiving, by the first server from the second server, attribute information of the topology road segment generated by the road segment management server, wherein the attribute information of the topology road segment generated by the road segment management server includes a length of the topology road segment or a road grade of the topology road segment; and

generating road-condition information of the target road segment by inputting at least the second driving characteristic-information of the second vehicle on the topology road segment and the attribute information of the topology road segment to a trained road-condition prediction model,

wherein the road-condition prediction model is configured to output a congestion state of the target road segment and is trained by:

obtaining first historical driving characteristic-information of a third vehicle passing the target road segment during a target period and second historical driving characteristic-information of a fourth vehicle passing the topology road segment during the target period;

obtaining a true road-condition probability value of the target road segment during the target period using the first historical driving characteristic-information and the second historical driving characteristic-information, including: matching the first historical driving characteristic-information and the second historical driving characteristic-information with an offline driving characteristic-information to obtain a matching result, and filtering non-driving information in the matching result to generate the true road-condition probability value; and

training the road-condition prediction model using a classification algorithm and based on the first historical driving characteristic-information, the second historical driving characteristic-information, and the true road-condition probability value,

wherein the road-condition prediction model comprises a congestion prediction model configured to output a first road-condition probability value and a clearness prediction model configured to output a second road-condition probability value, and the congestion state of the target road segment is determined based on the first road-condition probability value and the second road-condition probability value.

8. The method according to claim 1 , further comprising:

obtaining a true road-condition probability value based on the target road segment and the topology road segment in a target period; and

creating a road-condition prediction model based on the true road-condition probability, wherein the true road-condition probability value is in a one-to-one correspondence with the target period.

9. The method according to claim 1 , further comprising:

sending a real-time driving data obtaining request to a vehicle-mounted apparatus of the first vehicle or the second vehicle; and

receiving, from the vehicle-mounted apparatus, the first real-time driving data or the second real-time driving data.

10. The method according to claim 1 , wherein the first driving characteristic-information is obtained by:

collecting, using a camera, the first real-time driving data of the first vehicle; and

parsing the first real-time driving data to extract the first driving characteristic-information.

11. The method according to claim 1 , wherein the first driving characteristic-information includes a vehicle speed of the first vehicle on the target road segment, a stay duration of the first vehicle on the target road segment, and a number of traffic signal lights on the target road segment.

12. The method according to claim 1 , wherein the first vehicle and the second vehicle are the same vehicle.

13. The method according to claim 1 , further comprising:

inputting the first driving characteristic-information, the second driving characteristic-information, the first attribute information, and the second attribute information to the road-condition prediction model to obtain the road-condition information of the target road segment.

14. The method according to claim 1 , wherein:

the congestion state of the target road segment is determined based on a comparison between the first road-condition probability value and a congestion threshold and a comparison between the second road-condition probability value and a clearness threshold.

15. The method according to claim 1 , wherein:

determining that the target road segment to be in a clear state when the first road-condition probability value is less than a congestion threshold, and when the second road-condition probability value is greater than or equal to a clearness threshold;

determining that the target road segment to be in a slowing state when the first road-condition probability value is less than the congestion threshold, and when the second road-condition probability value is less than the clearness threshold; and

determining that the target road segment to be in a congested state when the first road-condition probability value is greater than or equal to the congestion threshold.