Granted Patent B2

US 12,697,978 · App. 18/205,944 · Granted Aug 4, 2026

Inventors: Wutu Lin (San Diego, CA); Xiaodi Hou (San Diego, CA)

Assignee: CreateAI, Inc.

B60W40/04B60W30/0956B60W60/0011G01C21/20G01S13/931G01S17/931G01S13/865G01S13/867

Abstract

Claims (42)

1 . A system comprising:

a data processor; and

a path planning module, executable by the data processor, the path planning module being configured to:

receive, from a remote data source, ancillary data to modify operation of the path planning module based on a context in which an autonomous vehicle is operating;

generate, by a modified operation of the path planning module, a first suggested trajectory comprising a plurality of waypoints;

generate, based on the first suggested trajectory, a trajectory gradient comprising scale and direction data corresponding to each of the plurality of waypoints;

modify, based on the trajectory gradient, parameters of a human driving model;

generate, based on the trajectory gradient and modified parameters of the human driving model, a second suggested trajectory; and

cause the autonomous vehicle to follow a path conforming to the second suggested trajectory.

2 . The system of claim 1 wherein the path planning module is configured to generate the first suggested trajectory using machine learnable components.

3 . The system of claim 1 , the trajectory gradient is generated based on a mathematical derivative of a function defining the first suggested trajectory for the autonomous vehicle.

4 . The system of claim 1 , wherein the path planning module is configured to score the first suggested trajectory.

5 . The system of claim 1 wherein the path planning module is configured to retain information corresponding to human driving behaviors as mathematical or data representations.

6 . The system of claim 1 , wherein the path planning module is configured to output information indicative of the path to a vehicle control subsystem causing the autonomous vehicle to follow the path conforming to the second suggested trajectory.

7 . The system of claim 1 wherein the remote data source is in wireless data communication with the autonomous vehicle.

8 . The system of claim 1 , wherein the path planning module is configured to:

generate a first score for the first suggested trajectory;

generate a second score for the second suggested trajectory;

determine, prior to causing the autonomous vehicle to follow the path conforming to the second suggested trajectory, that a difference between the first score and the second score is within a score differential threshold.

9 . A method comprising:

executing a path planning module by use of a data processor;

receiving, from a remote data source by use of the data processor, ancillary data to modify operation of the path planning module based on a context in which an autonomous vehicle is operating;

generating, by use of the data processor, a first suggested trajectory comprising a plurality of waypoints;

generating, by use of the data processor, based on the first suggested trajectory, a trajectory gradient comprising scale and direction data corresponding to each of the plurality of waypoints;

modifying, based on the trajectory gradient, parameters of a human driving model;

generating, based on the trajectory gradient and modified parameters of the human driving model, a second suggested trajectory; and

causing the autonomous vehicle to follow a path conforming to the second suggested trajectory.

10 . The method of claim 9 , wherein the human driving model is trained with training data corresponding to human driving behaviors.

11 . The method of claim 9 , wherein the trajectory gradient is generated based on a mathematical derivative of a function defining the first suggested trajectory for the autonomous vehicle.

12 . The method of claim 9 , including scoring the first suggested trajectory.

13 . The method of claim 9 , including causing the autonomous vehicle to follow a different path if a score corresponding to the first suggested trajectory is not within a score differential threshold.

14 . A non-transitory machine-useable storage medium embodying instructions which, when executed by a machine, cause the machine to:

execute a path planning module by use of a data processor;

receive, from a remote data source, ancillary data to modify operation of the path planning module based on a context in which an autonomous vehicle is operating;

generate, by a modified operation of the path planning module, a first suggested trajectory comprising a plurality of waypoints;

generate, based on the first suggested trajectory, a trajectory gradient comprising scale and direction data corresponding to each of the plurality of waypoints;

modify, based on the trajectory gradient, parameters of a human driving model;

generate, based on the trajectory gradient and modified parameters of the human driving model, a second suggested trajectory; and

cause the autonomous vehicle to follow a path conforming to the second suggested trajectory.

15 . The non-transitory machine-useable storage medium of claim 14 , wherein the trajectory gradient is generated based on a mathematical derivative of a function defining the first suggested trajectory for the autonomous vehicle.

16 . The non-transitory machine-useable storage medium of claim 14 , wherein the human driving model is trained with training data corresponding to human driving behaviors.

17 . The non-transitory machine-useable storage medium of claim 14 , wherein the path planning module is configured to output information indicative of the path to a vehicle control subsystem causing the autonomous vehicle to follow the path conforming to the second suggested trajectory.

Assignments (3)

Continuity (3)

References Cited (173)