System and method for predicting behaviors of detected objects through environment representation

Patent metadata

Field	Value
Publication number	US-9766626-B1
Application number	US-201615161556-A
Country	US
Kind code	B1
Filing date	May 23, 2016
Priority date	Feb 6, 2012
Publication date	Sep 19, 2017
Grant date	Sep 19, 2017

Abstract

Aspects of the invention relate generally to autonomous vehicles. The features described improve the safety, use, driver experience, and performance of these vehicles by performing a behavior analysis on mobile objects in the vicinity of an autonomous vehicle. Specifically, the autonomous vehicle is capable of detecting nearby objects, such as vehicles and pedestrians, and is able to determine how the detected vehicles and pedestrians perceive their surroundings. The autonomous vehicle may then use this information to safely maneuver around all nearby objects.

First claim

The invention claimed is:

1. A method for autonomous control of a first vehicle having one or more processors configured to control the first vehicle in an autonomous driving mode, the method comprising:
receiving, from an object sensing system of the vehicle including one or more sensors, information identifying a plurality of mobile objects external to the first vehicle as well as position and movement of each of the plurality of mobile objects;
identifying, by one or more processors, an object from the plurality of mobile objects corresponding to a second vehicle;
determining, by the one or more processors, how the first vehicle would behave if the first vehicle were placed in the position of the object;
using, by the one or more processors, the determination to predict a likely future behavior of the object; and
controlling, by the one or more processors, the first vehicle in the autonomous driving mode based on the predicted likely future behavior of the first object.

2. The method of claim 1 , wherein the object is a second vehicle, the position of the second vehicle indicates that the second vehicle is in a turning lane, and wherein the predicted likely future behavior of the second vehicle is turning at an intersection.

3. The method of claim 2 , wherein controlling the first vehicle includes slowing the first vehicle down as the first vehicle approaches the intersection.

4. The method of claim 1 , further comprising:
predicting a likely future behavior of a second object of the plurality of moving objects, wherein the second object is a an object other than a vehicle, using an object-type based prediction model, the movement of the second object, and the position of the second object; and
wherein controlling the vehicle is further based on the predicted likely future behavior of the second object.

5. The method of claim 4 , further comprising, prior to predicting the likely future behavior of the second object, determining that a type of the second object is a type other than a vehicle.

6. The method of claim 5 , further comprising, after determining that the second object is the type other than a vehicle, selecting the object-type based prediction model from a plurality of object-type based prediction model each associated with an object type, such that the associated object type of the selected object-type based prediction model corresponds to the type other than a vehicle.

7. The method of claim 4 , wherein the second object is a bicycle, and the position of the bicycle indicates that the bicycle is beginning to ascend a hill in front of the vehicle, and wherein the predicted likely future behavior of the bicycle is slowing down.

8. The method of claim 7 , wherein controlling the first vehicle includes slowing the first vehicle down regardless of the acceleration of the bicycle.

9. The method of claim 4 , wherein the second object is a pedestrian, and the position of the pedestrian indicates that the pedestrian is approaching a crosswalk, and wherein the predicted likely future behavior of the pedestrian is crossing a road using the crosswalk.

10. The method of claim 9 , wherein predicting a likely future behavior of the second object is further based on the position and movement of at least one other object of the plurality of moving objects.

11. The method of claim 10 , wherein the at least one other object of the plurality of moving objects is a vehicle passing through the crosswalk, and wherein the predicted likely future behavior of the pedestrian is crossing the road using the crosswalk after the vehicle has passed through the crosswalk.

12. The method of claim 1 , wherein determining how the first vehicle would behave is further based on the position and movement of other objects of the plurality of moving objects.

13. The method of claim 12 , wherein the object is a second vehicle, one of the other objects is a third vehicle in front of the second vehicle and making a turn, and the predicted likely future behavior of the third vehicle is changing to an adjacent lane.

14. The method of claim 13 , wherein controlling the first vehicle includes not changing to the adjacent lane.

15. The method of claim 12 , wherein the object is a second vehicle and the other objects of the plurality of moving objects includes at least two additional vehicles one of which is stopped, and the predicted likely future behavior of the second vehicle is stopping.

16. The method of claim 15 , wherein controlling the first vehicle includes stopping the first vehicle before making a turning maneuver.

17. A system for controlling a first vehicle, the system comprising:
one or more processors configured to:
receive, from an object sensing system of the vehicle including one or more sensors, information identifying a plurality of mobile objects external to the first vehicle as well as position and movement of each of the plurality of mobile objects;
identify an object from the plurality of mobile objects corresponding to a second vehicle;
determine how the first vehicle would behave if the first vehicle were placed in the position of the object;
use the determination to predict a likely future behavior of the object; and
control the first vehicle based on the predicted likely future behavior of the first object.

18. The system of claim 17 , wherein the one or more processors are further configured to:
predict a likely future behavior of a second object of the plurality of moving objects, wherein the second object is a an object other than a vehicle, using an object-type based prediction model, the movement of the second object, and the position of the second object; and
control the vehicle further based on the predicted likely future behavior of the second object.

19. The system of claim 18 , wherein the one or more processors are further configured to, prior to predicting the likely future behavior of the second object, determine that a type of the second object is a type other than a vehicle.

20. The system of claim 17 , further comprising the vehicle.

Assignees

• Waymo Llc

Inventors

• Zhu Jiajun
• Ferguson David I

Classifications

• G08G1/096791

  where the origin of the information is another vehicle · CPC title

• G01S17/931

  of land vehicles · CPC title

• G08G1/166

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

• G01S2013/932

  using own vehicle data, e.g. ground speed, steering wheel direction · CPC title

• G01S2013/93273

  on the top of the vehicles · CPC title