Behavior prediction system

Patent metadata

Field	Value
Publication number	US-11377100-B2
Application number	US-201916700818-A
Country	US
Kind code	B2
Filing date	Dec 2, 2019
Priority date	Dec 19, 2018
Publication date	Jul 5, 2022
Grant date	Jul 5, 2022

Abstract

A behavior prediction system includes an information obtainer, a probability distribution model generator, a map generator, and a display. The information obtainer obtains information on a traveling direction of a moving object whose behavior is to be predicted. The probability distribution model generator generates a probability distribution model regarding a traveling direction of the moving object based on the information on an obtained traveling direction of the moving object. The map generator uses the probability distribution model to calculate probabilities that the moving object passes through respective areas into which a movable region of the moving object is divided and that generates a probability map in which the probabilities are assigned to the respective areas. The display displays the probability map.

First claim

What is claimed is:

1. A behavior prediction system comprising:
a processor; and
a display,
wherein the processor is programmed to perform as:
an information obtainer that obtains information on a traveling direction of a moving object whose behavior is to be predicted;
a probability distribution model generator that generates a probability distribution model regarding a traveling direction of the moving object based on a movable region of the moving object and the information on an obtained traveling direction of the moving object; and
a map generator that uses the probability distribution model to calculate probabilities that the moving object passes through respective areas into which the movable region of the moving object is divided and that generates a probability map in which the probabilities are assigned to the respective areas, the probability map having a same shape as that of the movable region,

wherein the probability distribution model generator generates the probability distribution model based on data on the traveling direction of the moving object,
wherein the data includes first data calculated based on first information on the traveling direction of the moving object and second data calculated based on second information of the traveling direction of the moving object, the second information being obtained by the information obtainer subsequent to the first information, and
wherein the display displays the probability map.

2. The behavior prediction system according to claim 1 , wherein
the probability distribution model generator generates the probability distribution model based on the information on the obtained traveling direction of the moving object according to directional statistics.

3. The behavior prediction system according to claim 2 , wherein
the probability distribution model generator generates the probability distribution model using von Mises distribution.

4. The behavior prediction system according to claim 2 , wherein,
in response to change in tendency of a traveling direction of the moving object, the probability distribution model generator resets the data on the traveling direction of the moving object calculated before the change based on the information on the traveling direction of the moving object, and generates the probability distribution model based on data on a newly calculated traveling direction of the moving object.

5. The behavior prediction system according to claim 2 , wherein,
the map generator calculates probabilities that the moving object passes through the respective areas by a dynamic programming method.

6. The behavior prediction system according to claim 1 , wherein
the probability distribution model generator generates the probability distribution model using von Mises distribution.

7. The behavior prediction system according to claim 6 , wherein,
in response to change in tendency of a traveling direction of the moving object, the probability distribution model generator resets the data on the traveling direction of the moving object calculated before the change based on the information on the traveling direction of the moving object, and generates the probability distribution model based on data on a newly calculated traveling direction of the moving object.

8. The behavior prediction system according to claim 6 , wherein,
the map generator calculates probabilities that the moving object passes through the respective areas by a dynamic programming method.

9. The behavior prediction system according to claim 1 , wherein,
in response to change in tendency of a traveling direction of the moving object, the probability distribution model generator resets the data on the traveling direction of the moving object calculated before the change based on the information on the traveling direction of the moving object, and generates the probability distribution model based on data on a newly calculated traveling direction of the moving object.

10. The behavior prediction system according to claim 1 , wherein,
the map generator calculates probabilities that the moving object passes through the respective areas by a dynamic programming method.

11. The behavior prediction system according to claim 10 , wherein
the map generator uses the probability distribution model to calculate probabilities that the moving object passes through the respective areas, while calculating a turning rate of the moving object and modifying the probability distribution model based on the turning rate.

12. The behavior prediction system according to claim 1 , wherein the system provides another moving object with the probability map (1) to capture the moving object, (2) to avoid colliding with the moving object based on the probability map, or (3) to following the moving object.

13. The behavior prediction system according to claim 12 , wherein the probability distribution model generator generates the probability distribution model using von Mises distribution.

14. The behavior prediction system according to claim 12 , wherein, in response to change in tendency of a traveling direction of the moving object, the probability distribution model generator resets the data on the traveling direction of the moving object calculated before the change based on the information on the traveling direction of the moving object, and generates the probability distribution model based on data on a newly calculated traveling direction of the moving object.

15. The behavior prediction system according to claim 12 , wherein, the map generator calculates probabilities that the moving object passes through the respective areas by a dynamic programming method.

Assignees

• Subaru Corp

Inventors

• Itabashi Naoaki
• Kono Mitsuru
• Hoshino Satoshi
• Katsumoto Takeshi

Classifications

• G08G5/80

  Anti-collision systems · CPC title

• G08G5/21

  located onboard the aircraft · CPC title

• G08G5/25

  Transmission of traffic-related information between aircraft · CPC title

• G08G1/166

  for active traffic, e.g. moving vehicles, pedestrians, bikes · CPC title

• G06T7/251

  involving models · CPC title