Method, apparatus and electronic device for early-warning

Patent metadata

Field	Value
Publication number	US-11645909-B2
Application number	US-202017034360-A
Country	US
Kind code	B2
Filing date	Sep 28, 2020
Priority date	Jun 29, 2020
Publication date	May 9, 2023
Grant date	May 9, 2023

Abstract

The present disclosure discloses a method, apparatus and electronic device for early-warning, relates to the field of internet of vehicle and used for an intelligent traffic scenario. The specific implementations are: determining a target object crossing a road at a current moment, the road crossed by the target object being a first road; determining a target area based on a location of the target object; determining a target vehicle on the first road based on locations of vehicles on the first road as well as the target area; and sending an early-warning message to the target vehicle. It is using the locations of vehicles on the first road as well as the location of the target object crossing the road to determine the target vehicle during determination of the target vehicle among the vehicles on the first road, in order to send early-warning message to the target vehicle to achieve early-warning of the target vehicle and increase precision of the early-warning.

First claim

What is claimed is:

1. An early-warning method, comprising:
determining a target object crossing a road at a current moment, the road crossed by the target object being a first road;
determining a target area based on a location of the target object;
determining a target vehicle on the first road based on locations of vehicles on the first road as well as the target area; and
sending an early-warning message to the target vehicle,
wherein the determining the target area based on the location of the target object comprises:
determining the target area corresponding to a first predetermined time period and having the location of the target object as a center of the target area, on the basis of the location of the target object, and
wherein the determining the target vehicle on the first road based on locations of vehicles on the first road as well as the target area comprises:
determining the target vehicle out of respective vehicles on the first road and within the target area, on the basis of locations of the respective vehicles on the first road as well as the target area.

2. The method according to claim 1 , wherein determining the target object crossing the road at the current moment comprises:
determining a road on which candidate objects are located at the current moment as well as travelling directions of the candidate objects at the current moment; and
determining the target object among the candidate objects based on angles between the travelling directions of the candidate objects and a direction of the road on which the candidate objects are located.

3. The method according to claim 2 , wherein determining the road on which the candidate objects are located at the current moment as well as travelling directions of the candidate objects at the current moment comprises:
acquiring a first location of a first candidate object at the current moment by means of a V2I communication between vehicles and the road, and acquiring a second location of a second candidate object by means of target detection;
acquiring locations of the candidate objects at the current moment by merging the first location of the first candidate object with the second location of the second candidate object;
calculating travelling directions of the candidate objects at the current moment based on the locations of the candidate objects at the current moment as well as locations of the candidate objects at a first moment before and adjacent to the current moment; and
determining roads on which the candidate objects are located based on the locations of the candidate objects as well as map data.

4. The method according to claim 1 , wherein determining the target vehicle on the first road based on locations of the vehicles on the first road as well as the target area comprises:
calculating target locations reached by the vehicles on the first road after the first predetermined time period based on the locations of the vehicles on the first road, the first predetermined time period and driving speed of the vehicles on the first road at the current moment; and
determining the target vehicle among the vehicles on the first road based on distances between the target locations of the vehicles on the first road and the target area.

5. The method according to claim 1 , wherein after determining the target area based on the location of the target object and before sending the early-warning message to the target vehicle, the method further comprises:
in case of the location of the target object is a junction location and color of traffic light of the first road is not a first predetermined color, or the location of the target object is not the junction location, generating the early-warning message based on the target object and the first predetermined time period.

6. The method according to claim 1 , wherein the early-warning method is realized by a computer device.

7. An electronic device, comprising:
at least one processor; and
a memory in communication with the at least one processor, wherein instructions which can be processed by the at least one processor are stored in the memory such that the at least one processor carries out an early-warning method, comprising:
determining a target object crossing a road at a current moment, the road crossed by the target object being a first road;
determining a target area based on a location of the target object;
determining a target vehicle on the first road based on locations of vehicles on the first road as well as the target area; and
sending an early-warning message to the target vehicle,
wherein the determining the target area based on the location of the target object comprises:
determining the target area corresponding to a first predetermined time period and having the location of the target object as a center of the target area, on the basis of the location of the target object, and
wherein the determining the target vehicle on the first road based on locations of vehicles on the first road as well as the target area comprises:
determining the target vehicle out of respective vehicles on the first road and within the target area, on the basis of locations of the respective vehicles on the first road as well as the target area.

8. The electronic device according to claim 7 , wherein determining the target object crossing the road at the current moment comprises:
determining a road on which candidate objects are located at the current moment as well as travelling directions of the candidate objects at the current moment; and
determining the target object among the candidate objects based on angles between the travelling directions of the candidate objects and a direction of the road on which the candidate objects are located.

9. The electronic device according to claim 8 , wherein determining the road on which the candidate objects are located at the current moment as well as travelling directions of the candidate objects at the current moment comprises:
acquiring a first location of a first candidate object at the current moment by means of a V2I communication between vehicles and the road, and acquiring a second location of a second candidate object by means of target detection;
acquiring locations of the candidate objects at the current moment by merging the first location of the first candidate object with the second location of the second candidate object;
calculating travelling directions of the candidate objects at the current moment based on the locations of the candidate objects at the current moment as well as locations of the candidate objects at a first moment before and adjacent to the current moment; and
determining roads on which the candidate objects are located based on the locations of the candidate objects as well as map data.

10. The electronic device according to claim 7 , wherein determining the target vehicle on the first road based on locations of the vehicles on the first road as well as the target area comprises:
calculating target locations reached by the vehicles on the first road after the first predetermined time period based on the locations of the vehicles on the first road, the first predetermined time period and driving speed of the vehicles on the first road at the current moment; and
determining the target vehicle among the vehicles on the first road based on distances between the target locations of the vehicles on the first road and the target area.

11. The electronic device according to claim 7 , wherein after determining the target area based on the location of the target object and before sending the early-warning message to the target vehicle, the method further comprises:
in case of the locati

Assignees

• Beijing Baidu Netcom Sci & Tech Co Ltd
• Beijing Baidu Netcom Sci & Tech Co Ltd

Inventors

• Xia Deguo
• Huang Jizhou
• Zhao Hui
• Zhang Liuhui
• Bai Hongxia
• Liu Yuting

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G08G1/164

  Centralised systems, e.g. external to vehicles · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G08G1/0116

  from roadside infrastructure, e.g. beacons · CPC title

• G08G1/0125

  Traffic data processing · CPC title