US 10,293,483 B2 · Apparatus and methods for training path navigation by robots

Abstract

An apparatus and methods for training and/or operating a robotic device to follow a trajectory. A robotic vehicle may utilize a camera and stores the sequence of images of a visual scene seen when following a trajectory during training in an ordered buffer. Motor commands associated with a given image may be stored. During autonomous operation, an acquired image may be compared with one or more images from the training buffer in order to determine the most likely match. An evaluation may be performed in order to determine if the image may correspond to a shifted (e.g., left/right) version of a stored image as previously observed. If the new image is shifted left, right turn command may be issued. If the new image is shifted right then left turn command may be issued.

Claims20

Dependent claims

A method of determining a control signal for a robot executed by at least one processor upon execution of computer readable instructions stored in a memory of the robot, comprising:

receiving a plurality of features, each of the plurality of features characterized by a plurality of types of features;
determining a subset of features based on a random configuration;
comparing individual features of the determined subset to a plurality of training feature sets, the plurality of training feature sets characterized by a number of training features;
determining a similarity measure based on the comparison;
determining one or more potential control signals for the robot upon the similarity measure reaching a threshold, a respective one of the one or more potential control signals being associated with a corresponding training set of the plurality of training feature sets;
selecting the control signal from the determined one or more potential control signals;
and
applying transformations to at least one image received from at least one sensor on the robot, the transformations including shifting and cropping to the at least one image.

5 more elements

A non-transitory computer-readable storage medium comprising a plurality of instructions stored thereon, the instructions, when executed by a processor, cause the processor to:

receive a plurality of features, each of the plurality of features characterized by a plurality of types of features;
determine a subset of features based on a random configuration;
compare individual features of the determined subset to a plurality of training feature sets, the plurality of training feature sets characterized by a number of training features;
determine a similarity measure based on the comparison;
determine one or more potential control signals for the robot upon the similarity measure reaching a threshold, a respective one of the one or more potential control signals being associated with a corresponding training set of the plurality of training feature sets;
select the control signal from the determined one or more potential control signals;
and
apply transformations to at least one image received from the at least one sensor on the robot, the transformations comprising shifting and cropping to the at least one image.

5 more elements

A robotic apparatus comprising:

at least one sensor configured to collect environment data;
at least one actuator configured to move the robotic apparatus;
a processor configured to execute computer readable instructions to:
receive a plurality of features, each of the plurality of features characterized by a plurality of types of features;
determine a subset of features based on a random configuration;
compare individual features of the determined subset to a plurality of training feature sets, the plurality of training feature sets characterized by a number of training features;
determine a similarity measure based on the comparison;
determine one or more potential control signals for the robot upon the similarity measure reaching a threshold, a respective one of the one or more potential control signals being associated with a corresponding training set of the plurality of training feature sets;
select the control signal from the determined one or more potential control signals.

6 more elements

Dependencies are inferred from the claim text.

Apparatus and methods for training path navigation by robots

Life of this patent

Apparatus and methods for training path navigation by robots

Life of this patent

Abstract

Claims20