Trajectory modifications based on a collision zone

Patent metadata

Field	Value
Publication number	US-2022363256-A1
Application number	US-202217870671-A
Country	US
Kind code	A1
Filing date	Jul 21, 2022
Priority date	Oct 24, 2019
Publication date	Nov 17, 2022
Grant date	—

Abstract

The described techniques relate to modifying a trajectory of a vehicle, such as an autonomous vehicle, based on an overlap area associated with an object in the environment. In examples, map data may be used, in part, to generate an initial trajectory for an autonomous vehicle to follow through an environment. In some cases, a yield trajectory may be generated based on detection of the object, and the autonomous vehicle may evaluate a cost function to determine whether to execute the yield or follow the initial trajectory. In a similar manner, the autonomous vehicle may determine a merge location of two lanes of a junction, and use the merge location to update extents of an overlap area to prevent the autonomous vehicle from blocking the junction and/or provide sufficient space to yield to the oncoming vehicle while merging.

First claim

What is claimed is:

1 . A system comprising:
one or more processors; and one or more computer-readable media storing instructions that, when executed by the one or more processors, perform operations comprising:
detecting, from sensor data, an object in the environment;
determining a path associated with the object;
determining an area in the environment based at least in part on the path and a set of intended locations of a vehicle traversing the environment;
determining, based at least in part on determining whether the vehicle yielding to the object will block the path of the object and the area, a cost;
determining, based at least in part on the cost, a command; and
controlling the vehicle in accordance with the command.

2 . The system of claim 1 , wherein controlling the vehicle in accordance with the command comprises one of:
controlling the vehicle to proceed through the set of intended locations, or controlling the vehicle to proceed through the set of intended locations while yielding to the object.

3 . The system of claim 1 , wherein determining the cost is based at least in part on an overlap associated with the set of intended locations and the path, the overlap comprising at least one of:
a spatially based overlap determined based at least in part on first predicted locations associated with the vehicle entering and exiting the area and second predicted locations associated with the object entering and exiting the area; and a temporally based overlap based at least in part on first predicted times associated with the first predicted locations and second predicted times associated with the second predicted locations.

4 . The system of claim 3 , wherein the second predicted times are based at least in part on determining one or more of:
an object type of the object; a current speed of the object; a junction type of a junction where the overlap area is located; a speed limit at or proximate to the junction; or a predicted rate of acceleration or a rate of deceleration of the object.

5 . The system of claim 3 , wherein the cost is further based on determining a level of safety based at least in part on a likelihood of overlap of the vehicle and the object by comparing the first predicted times and the second predicted times.

6 . The system of claim 1 , wherein the cost is based at least in part on a distance component associated with a minimum distance to stop to perform the yield by the vehicle into a junction and a length of an overlap zone associated with the area, the overlap zone comprising an area of a possible overlap of the vehicle and the object.

7 . A method comprising:
determining a predicted path of an object in an environment; determining a yield to be performed by the vehicle based at least in part on the predicted path of the object; determining whether to cause the vehicle to perform the yield based at least in part on:
determining an area in the environment based at least in part on a set of intended locations of the vehicle and the predicted path; and
determining, based at least in part on the area, a cost associated with performing the yield; and
controlling the vehicle based at least in part on the cost.

8 . The method of claim 7 , wherein the predicted path is one of multiple possible predicted paths for the object, the method further comprising determining that the predicted path and a variance associated with the predicted path is more conservative than at least one other path of the multiple possible predicted paths.

9 . The method of claim 8 , further comprising determining, based at least in part on the predicted path being more conservative than at least one other path of the multiple possible predicted paths, an entry point and an exit location for the overlap area based at least in part on the variance.

10 . The method of claim 7 , further comprising:
determining an overlap associated with the area, wherein the overlap is based at least in part on:
a spatial overlap based at least in part on first predicted locations associated with the vehicle entering and exiting the area and second predicted locations associated with the object entering and exiting the area; and
a temporally based overlap based at least in part on first predicted times associated with the first predicted locations and second predicted times associated with the second predicted locations,
wherein the cost is based at least in part on the overlap.

11 . The method of claim 10 , wherein the overlap is further based on a potential variance of the path.

12 . The method of claim 10 , wherein the overlap is further based on a potential variance of a tracked speed of the object.

13 . The method of claim 7 , further comprising:
determining multiple paths associated with the object in the environment, wherein a path of the multiple paths corresponds to respective predictions of paths that the object will follow, the path of the object being one of the multiple paths; and determining a safety envelope associated with controlling the vehicle, wherein the safety envelope prevents an overlap between the vehicle and the object.

14 . The method of claim 13 , wherein determining the safety envelope comprises determining one or more variances for an individual path of the multiple paths, a variance of the one or more variances being based on a certainty of an individual path of the multiple paths, a first error associated with a speed of the object, or a second error associated with an acceleration of the object.

15 . The method of claim 7 , wherein controlling the vehicle to proceed along the trajectory prevents the vehicle from blocking the path of the object.

16 . One or more computer-readable media storing instructions that, when executed by one or more processors, perform operations comprising:
determining a predicted path of an object in the environment; determining a yield to be performed by a vehicle based at least in part on the predicted path of the object; determining whether to cause the vehicle yield to the object based at least in part on:
determining an area in the environment the vehicle following an intended trajectory and the predicted path; and
determining a cost associated with performing the yield, the cost based at least in part on the area; and
controlling the vehicle to proceed along the intended trajectory based at least in part on the cost.

17 . The one or more computer-readable media of claim 16 , wherein the cost is based at least in part on an overlap associated with the trajectory and the predicted path, the overlap comprising at least one of:
a spatial overlap based at least in part on first predicted locations associated with the vehicle and the area and second predicted locations associated with the object and the area; and a time overlap based at least in part on first predicted times and the first predicted locations and second predicted times for the object and the second predicted locations.

18 . The one or more computer-readable media of claim 17 , wherein the second predicted times are based at least in part on determining one or more of:
an object type of the object; a current speed of the object; a junction type of a junction where the overlap area is located; a speed limit at or proximate to the junction; or a predicted rate of acceleration or a rate of deceleration of the object.

19 . The one or more computer-readable media of claim 17 , wherein the first predicted locations are based at least in part

Assignees

• Zoox Inc

Inventors

• Dax Victoria Magdalena

Classifications

• G08G1/166 Primary

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

• B60W30/18154 Primary

  Approaching an intersection · CPC title

• B60W30/09

  Taking automatic action to avoid collision, e.g. braking and steering · CPC title

• B60W60/0011

  involving control alternatives for a single driving scenario, e.g. planning several paths to avoid obstacles · CPC title

• B60W30/0956

  the prediction being responsive to traffic or environmental parameters · CPC title