Vision-guided robots and methods of training them
First Claim
Patent Images
1. A robot-implemented method of manipulating objects based on visual recognition thereof, the method comprising causing a robot to execute steps comprising:
- (a) selecting an object in a camera field of view;
(b) using a visual model to computationally identify at least a class of the object based on visual detection of general characteristics associated with the object class by prior training;
(c) automatically determining whether an object-specific manipulation routine is stored in association with the stored representation; and
(d) if so, using a controller to execute the object-specific manipulation routine, and if not, using a controller to execute a generic manipulation routine according to the object class.
1 Assignment
0 Petitions
Accused Products
Abstract
Via intuitive interactions with a user, robots may be trained to perform tasks such as visually detecting and identifying physical objects and/or manipulating objects. In some embodiments, training is facilitated by the robot'"'"'s simulation of task-execution using augmented-reality techniques.
126 Citations
3 Claims
-
1. A robot-implemented method of manipulating objects based on visual recognition thereof, the method comprising causing a robot to execute steps comprising:
-
(a) selecting an object in a camera field of view; (b) using a visual model to computationally identify at least a class of the object based on visual detection of general characteristics associated with the object class by prior training; (c) automatically determining whether an object-specific manipulation routine is stored in association with the stored representation; and (d) if so, using a controller to execute the object-specific manipulation routine, and if not, using a controller to execute a generic manipulation routine according to the object class.
-
-
2. A vision-guided robot for manipulating objects based on visual recognition thereof, the robot comprising:
-
at least one movable appendage for manipulating objects; a camera for capturing an image within a camera field of view; and a controller for (i) computationally identifying at least a class of an object in the image based on visual detection of general characteristics associated with the object class by prior training, (ii) automatically determining whether an object-specific manipulation routine is stored in association with the stored representation, and (iii) if so, causing the at least one movable appendage to execute the object-specific manipulation routine, and if not, causing the at least one movable appendage to execute a generic manipulation routine according to the object class. - View Dependent Claims (3)
-
Specification