Station recommendation

Patent metadata

Field	Value
Publication number	US-12140441-B2
Application number	US-202117531529-A
Country	US
Kind code	B2
Filing date	Nov 19, 2021
Priority date	Dec 25, 2020
Publication date	Nov 12, 2024
Grant date	Nov 12, 2024

Abstract

A method for recommending a station for a vehicle, a device, and a storage medium are provided. The method comprises: receiving, by a server, an access request from a vehicle; obtaining, based on the access request, a plurality of observation values from a plurality of stations associated with the vehicle, respectively, each observation value is based on a corresponding pre-trained recommendation model, each observation value includes factors associated with access of the vehicle to the station corresponding to the observation value; determining, an action value for the station based on the observation value and the pre-trained recommendation model for the station, the action value for the station indicates a matching degree between the access request and the station; determining a recommended station among the plurality of stations based on the action values of the plurality of stations; and sending to the vehicle an instruction of driving to the recommended station.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a server, an access request from a vehicle;
obtaining, by the server based on the access request, a plurality of observation values from a plurality of stations associated with the vehicle, respectively, wherein each observation value is based on a corresponding pre-trained recommendation model, and wherein each observation value comprises factors associated with access of the vehicle to the station corresponding to the observation value, wherein, for each station of the plurality of stations, the observation value comprises a first observation value for the station and second observation values for other stations within a second distance threshold from the station, wherein, for each station of the plurality of stations, obtaining the observation value comprises:
determining a corresponding weight for each of the second observation values according to degrees of association between the other stations and the station, wherein the plurality of stations share a same recommendation model, and differences between the plurality of stations are reflected in different observation values of the recommendation model, and wherein after one of the plurality of stations learns, the recommendation model is updated, and the updated recommendation model is applicable to the plurality of stations;

determining, by the server, for each station of the plurality of stations, an action value for the station based on the observation value and the corresponding pre-trained recommendation model for the station, wherein the action value for the station indicates a matching degree between the access request and the station;
determining, by the server, a recommended station among the plurality of stations based on the action values of the plurality of stations; and
sending, by the server to the vehicle, an instruction of driving to the recommended station.

2. The method according to claim 1 , wherein, for each station, the corresponding pre-trained recommendation model is trained based on a historical recommendation and a reward value corresponding to the historical recommendation.

3. The method according to claim 2 , wherein the reward value is determined based on one or more of: access waiting time of the vehicle at the station, a reward threshold, expense of the vehicle at the station, a matching degree between a model of the vehicle and an accessible model for the station, station access efficiency of the station, or a score of the station provided by the vehicle.

4. The method according to claim 3 , further comprising:
obtaining a first time point as a time point of sending the instruction to the vehicle;
determining that the vehicle accessed the recommended station within a time threshold starting from the first time point;
obtaining a second time point that is a time point when the vehicle accessed the recommended station;
determining the access waiting time of the vehicle at the recommended station based on the first time point and the second time point;
determining the reward value based on the access waiting time; and
associating the reward value with the historical recommendation.

5. The method according to claim 3 , further comprising:
obtaining a first time point as a time point of sending the instruction to the vehicle;
determining that the vehicle did not access the recommended station within a time threshold starting from the first time point;
determining the reward value based on the reward threshold; and
associating the reward value with the historical recommendation.

6. The method according to claim 1 , wherein the plurality of stations are stations within a first distance threshold from the vehicle.

7. The method according to claim 3 , wherein the reward value is determined based on the access waiting time and a reward value associated with the first time point.

8. The method according to claim 1 , wherein determining the recommended station comprises:
determining the action value of each of the plurality of stations;
determining a largest action value from the action values; and
determining a station corresponding to the largest action value as the recommended station.

9. The method according to claim 1 , wherein, for each station, the observation value indicates:
a current time;
a number of vacant access points for the station; and
driving time from a location where the access request is sent to the station.

10. The method according to claim 9 , wherein, for each station, the observation value further indicates access efficiency of the station.

11. The method according to claim 1 , wherein each station of the plurality of stations is a charging station.

12. An electronic device, comprising:
at least one processor; and
a memory in communication with the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to:
receive an access request from a vehicle;
obtain, based on the access request, a plurality of observation values from a plurality of stations associated with the vehicle, respectively, wherein each observation value is based on a corresponding pre-trained recommendation model, and wherein each observation value comprises factors associated with access of the vehicle to the station corresponding to the observation value, wherein, for each station of the plurality of stations, the observation value comprises a first observation value for the station and second observation values for other stations within a second distance threshold from the station, wherein, for each station of the plurality of stations, obtaining the observation value comprises:
determining a corresponding weight for each of the second observation values according to degrees of association between the other stations and the station, wherein the plurality of stations share a same recommendation model, and differences between the plurality of stations are reflected in different observation values of the recommendation model, and wherein after one of the plurality of stations learns, the recommendation model is updated, and the updated recommendation model is applicable to the plurality of stations;
determine, for each station of the plurality of stations, an action value for the station based on the observation value and the corresponding pre-trained recommendation model for the station, wherein the action value indicates a matching degree between the access request and the station;
determine a recommended station among the plurality of stations based on the action values of the plurality of stations; and
send to the vehicle an instruction of driving to the recommended station.

13. The electronic device of claim 12 , wherein, for each station, the pre-trained recommendation model is trained based on a historical recommendation and a reward value corresponding to the historical recommendation.

14. The electronic device according to claim 13 , wherein the reward value is determined based on one or more of: access waiting time of the vehicle at the station, a reward threshold, expense of the vehicle at the station, a matching degree between a model of the vehicle and an accessible model for the station, station access efficiency of the station, or a score of the station provided by the vehicle.

15. The electronic device according to claim 14 , wherein the instructions, when executed by the at least one processor, further enable the at least one processor to:
obtain a first time point as a time point of sending the instruction to the vehicle;
determine that the vehicle acces

Assignees

• Beijing Baidu Netcom Sci & Tech Co Ltd

Inventors

• Zhang Weijia
• Liu Hao
• Dou Dejing
• Xiong Hui

Classifications

• H04L67/12

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title

• G06Q30/0631 Primary

  Recommending goods or services · CPC title

• G01C21/3484

  Personalized, e.g. from learned user behaviour or user-defined profiles · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N3/006

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title