US 11,693,415 B2Granted

Predicting cut-in probabilities of surrounding agents

AssigneeWaymo LLC

InventorsKhaled Refaat, Chi Pang Lam

Life of this patent

Estimating the term…

Loading patent…

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating cut-in probabilities of agents surrounding a vehicle. One of the methods includes obtaining agent trajectory data for one or more agents in an environment; obtaining vehicle trajectory data of a vehicle in the environment; and processing a network input generated from the agent trajectory data and vehicle trajectory data using a neural network to generate a cut-in output, wherein the cut-in output comprises respective cut-in probabilities for each of a plurality of locations in the environment, wherein the respective cut-in probability for each location that is a current location of one of the one or more agents characterizes a likelihood that the agent in the current location will intersect with a planned future location of the vehicle within a predetermined amount of time.

Claims19

Dependent claims

A method comprising:

obtaining agent trajectory data comprising, for each particular agent of one or more agents in an environment:
for a current time point, a current location of the particular agent and current values for a predetermined set of motion parameters of the particular agent;
and
for each of one or more previous time points, a respective previous location of the particular agent and respective previous values for the predetermined set of motion parameters of the particular agent;
obtaining vehicle trajectory data of a vehicle in the environment, the vehicle trajectory data comprising:
for the current time point, a current location of the vehicle and current values for the predetermined set of motion parameters of the vehicle;
for each of the one or more previous time points, a respective previous location of the vehicle and respective previous values for the predetermined set of motion parameters of the vehicle;
and
for each of one or more future time points, a respective planned future location of the vehicle and respective planned future values for the predetermined set of motion parameters of the vehicle;
and
processing a network input generated from the agent trajectory data and vehicle trajectory data using a neural network to generate a single cut-in output, wherein the single cut-in output comprises a respective cut-in probability for each of a plurality of locations in the environment, the plurality of locations comprising one or more agent locations that are each a current location of a different one of the one or more agents and non-agent locations that are not any of the current locations of the one or more agents,
wherein, for each particular agent location, the respective cut-in probability characterizes a predicted likelihood that the particular agent in the particular agent location at the current time point will intersect with at least one respective planned future location of the vehicle within a predetermined amount of time.

9 more elements

A system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

obtaining agent trajectory data comprising, for each particular agent of one or more agents in an environment:
for a current time point, a current location of the particular agent and current values for a predetermined set of motion parameters of the particular agent;
and
for each of one or more previous time points, a respective previous location of the particular agent and respective previous values for the predetermined set of motion parameters of the particular agent;
obtaining vehicle trajectory data of a vehicle in the environment, the vehicle trajectory data comprising:
for the current time point, a current location of the vehicle and current values for the predetermined set of motion parameters of the vehicle;
for each of the one or more previous time points, a respective previous location of the vehicle and respective previous values for the predetermined set of motion parameters of the vehicle;
and
for each of one or more future time points, a respective planned future location of the vehicle and respective planned future values for the predetermined set of motion parameters of the vehicle;
and
processing a network input generated from the agent trajectory data and vehicle trajectory data using a neural network to generate a single cut-in output, wherein the single cut-in output comprises a respective cut-in probability for each of a plurality of locations in the environment, the plurality of locations comprising one or more agent locations that are each a current location of a different one of the one or more agents and non-agent locations that are not any of the current locations of the one or more agents,
wherein, for each particular agent location, the respective cut-in probability characterizes a predicted likelihood that the particular agent in the particular agent location at the current time point will intersect with at least one respective planned future location of the vehicle within a predetermined amount of time.

9 more elements

One or more non-transitory computer-readable storage media storing instructions that when executed by one or more computers cause the one or more computers to perform operations comprising:

obtaining agent trajectory data comprising, for each particular agent of one or more agents in an environment:
for a current time point, a current location of the particular agent and current values for a predetermined set of motion parameters of the particular agent;
and
for each of one or more previous time points, a respective previous location of the particular agent and respective previous values for the predetermined set of motion parameters of the particular agent;
obtaining vehicle trajectory data of a vehicle in the environment, the vehicle trajectory data comprising:
for the current time point, a current location of the vehicle and current values for the predetermined set of motion parameters of the vehicle;
for each of the one or more previous time points, a respective previous location of the vehicle and respective previous values for the predetermined set of motion parameters of the vehicle;
and
for each of one or more future time points, a respective planned future location of the vehicle and respective planned future values for the predetermined set of motion parameters of the vehicle;
and
processing a network input generated from the agent trajectory data and vehicle trajectory data using a neural network to generate a single cut-in output, wherein the single cut-in output comprises a respective cut-in probability for each of a plurality of locations in the environment, the plurality of locations comprising one or more agent locations that are each a current location of a different one of the one or more agents and non-agent locations that are not any of the current locations of the one or more agents,
wherein, for each particular agent location, the respective cut-in probability characterizes a predicted likelihood that the particular agent in the particular agent location at the current time point will intersect with at least one respective planned future location of the vehicle within a predetermined amount of time.

9 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/676,379Filed 6 Nov 2019
Published: US 2021/0132619 A16 May 2021
Granted: 4 Jul 2023
Term ends: Not estimatedThe source records don’t support an estimate.

Classification18 codes

G05D 1/0221
With means for defining a desired trajectory: involving a learning process
G01C 21/34
Specially adapted for navigation in a road network: route searching; Route guidance
G05D 1/0088
Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots: characterized by the…
G06N 3/006
Artificial life, i.e. computing arrangements simulating life: based on simulated virtual individual or collective life forms, e.g. social…