Collective determination among autonomous vehicles

• US 10,816,972 B2
• Filed: 03/15/2017
• Issued: 10/27/2020
• Est. Priority Date: 03/15/2017
• Status: Active Grant

• Current Assignee: Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: Toyota Research Institute, Inc. (Toyota Motor Corporation)
• Inventors: Michalakis, Nikolaos, Mason, Julian M.
• Primary Examiner(s): Nguyen, Tan Q
• Application Number:
  US15/459,305
• Publication Number:
  US 20180268698A1
• Time in Patent Office: 1,322 Days
• Field of Search: 701 23
• CPC Class Codes: B60W 40/02 related to ambient conditions
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00
• CPC Class Codes: G08G 1/0112 from the vehicle, e.g. floating car data [FCD]
• CPC Class Codes: G08G 1/096791 where the origin of the information is another vehicle
• CPC Class Codes: G08G 1/162 event-triggered
• CPC Class Codes: G08G 1/22 Platooning, i.e. convoy of communicating vehicles

Abstract

Aspects of the disclosure provide a method for collectively determining an object by a group of vehicles. The method can include receiving sensor data indicating an object at a first vehicle of the group of vehicles communicating with each other, determining the object to reach a first conclusion based on the sensor data at the first vehicle, transmitting the sensor data and the first conclusion to second vehicles of the group of vehicles from the first vehicle, determining the object to reach a second conclusion at each second vehicle based on the sensor data, transmitting second conclusions from the respective second vehicles to the other second vehicles and the first vehicle, and determining the object based on the first and second conclusions at the first vehicle or the second vehicles.

Claims

1. A method, comprising:

receiving sensor data indicating an object at a first vehicle belonging to a group of vehicles communicating with each other;

determining the object to reach a first conclusion based on the sensor data;

selecting second vehicles of the group of vehicles based on computational resources available in the second vehicles, each vehicle of the group of vehicles having different computational resources;

transmitting the sensor data to the selected second vehicles in the group of vehicles, the second vehicles determining the object to reach respective second conclusions based on the sensor data;

receiving the second conclusions from respective second vehicles; and

determining the object based on the first and second conclusions,

wherein each of the first vehicle and the second vehicles uses a different algorithm to determine the first conclusion and the second conclusions, respectively,

wherein the selected second vehicles are selected based on having more computation resources than other vehicles in the group of vehicles.

2. The method of claim 1, wherein the second vehicles are also selected based on having lowest communication delays to the first vehicle.

3. The method of claim 1, wherein the second vehicles are selected based on having the same make as the first vehicle.

4. The method of claim 1, wherein determining the object based on the first and second conclusions includes:

associating a weight to each of the first and second conclusions; and

drawing a final conclusion based on the weighted first and second conclusions.

5. The method of claim 4, wherein the weight is determined based on one of,

computation resources of the vehicle drawing the respective conclusion,

a reputation of the vehicle drawing the respective conclusion, or

a combination of a first weight for a second conclusion determined by the first vehicle and a second weight determined by a second vehicle generating the second conclusion.

6. A method, comprising:

receiving, at a second vehicle, sensor data indicating an object and a first conclusion of determining the object based on the sensor data from a first vehicle, the first and second vehicles belonging to a group of vehicles communicating with each other, each vehicle in the group of vehicles having different computational resources, the second vehicle being selected for receipt of the sensor data based on computational resources of the second vehicle;

determining the object to reach a second conclusion based on the received sensor data; and

transmitting the second conclusion from the second vehicle to ones of the group of vehicles,

wherein the first conclusion and the second conclusion are determined using different algorithms, and

wherein the second vehicle is selected based on having more computation resources than other vehicles in the group of vehicles.

7. The method of claim 6, further comprising:

receiving a third conclusion of determining the object based on the sensor data from a third vehicle of the group of vehicles, the third vehicle determining the object to reach the third conclusion based on the sensor data received from the first vehicle; and

determining the object based on the first, second, and third conclusions at the second vehicle.

8. The method of claim 7, further comprising:

receiving third conclusions of determining the object based on the sensor data from third vehicles of the group of vehicles, the third vehicles determining the object to reach respective third conclusions based on the sensor data received from the first vehicle; and

determining the object based on the first, second, and third conclusions at the second vehicle.

9. The method of claim 8, wherein determining the object based on the first, second, and third conclusions at the second vehicle includes:

associating a weight to each of the first, second and third conclusions; and

drawing a final conclusion based on the weighted first, second, and third conclusions.

10. The method of claim 9, wherein the weight is determined based on one of,

computation resources of the vehicle drawing the respective conclusion,

a reputation of the vehicle drawing the respective conclusion, or

a combination of a first weight for a second conclusion determined by the first vehicle and a second weight determined by a second vehicle generating the second conclusion.

11. A method, comprising:

receiving sensor data indicating an object at a first vehicle of a group of vehicles communicating with each other;

determining the object to reach a first conclusion based on the sensor data at the first vehicle;

selecting second vehicles of the group of vehicles based on computational resources available in the second vehicles, each vehicle in the group of vehicles having different computational resources;

transmitting the sensor data and the first conclusion to the selected second vehicles of the group of vehicles from the first vehicle;

determining the object to reach a second conclusion at each second vehicle based on the sensor data;

transmitting second conclusions from the respective second vehicles to the other second vehicles and the first vehicle; and

determining the object based on the first and second conclusions at the first vehicle or the second vehicles,

wherein the first conclusion and the second conclusions are determined using different algorithms, and

wherein the selected second vehicles are selected based on having more computation resources than other vehicles in the group of vehicles.

12. The method of claim 11, wherein the second vehicles are also selected based on having lowest communication delays to the first vehicle.

13. The method of claim 11, wherein the second vehicles are selected based on having the same make as the first vehicle.

14. The method of claim 11, wherein determining the object based on the first and second conclusions at the first vehicle or the second vehicles includes:

associating a weight to each of the first and second conclusions; and

drawing a final conclusion based on the weighted first and second conclusions.

15. The method of claim 14, wherein the weight is determined based on one of,

computation resources of the vehicle drawing the respective conclusion,

a reputation of the vehicle drawing the respective conclusion, or

a combination of a first weight for a second conclusion determined by the first vehicle and a second weight determined by a second vehicle generating the second conclusion.