Back to results
US 8,095,237 B2Granted
Method and apparatus for single image 3D vision guided robotics
AssigneeRoboticVISIONTech LLC
InventorsBabak Habibi, Simona Pescaru, Mohammad Sameti, Remus Florinel Boca
Life of this patent
Estimating the term…
Abstract
A method of three-dimensional object location and guidance to allow robotic manipulation of an object with variable position and orientation using a sensor array which is a collection of one or more sensors capable of forming a single image.
Claims28
A useful in three-dimensional pose estimation for use with a single camera mounted to a movable portion of a robot, the method comprising:
- capturing a two-dimensional image of a volume containing a target object;
- locating a number of features in the captured image of the target object;
- and
- determining by a processor an object space-to-camera space transformation for the target object based at least in part on a position of at least some of the located features using only the single captured image and an algorithm that employs a known or determinable physical relationship between at least some of the located features.
1 more element
An apparatus useful in robotics, the apparatus comprising:
- a single camera operable to capture a number of images of a calibration object;
- means for calibrating the camera, by:
- determining a set of intrinsic parameters of the camera from at least one of the number of images of the calibration object captured by the camera;
- and
- determining a set of extrinsic parameters of the camera from at least one of the number of images of the calibration object captured by the camera, the set of extrinsic parameters comprising a camera space-to-training space transformation defining a transformation between a camera space reference frame and a training space reference frame;
- and
- means for estimating a pose of a target object, by:
- capturing a two-dimensional image of a volume containing a target object;
- and
- locating at least six features in the captured image of the target object;
- and
- determining an object space-to-camera space transformation based at least in part on a position of at least some of the located features in solely the captured image using an algorithm that employs a known or determinable physical relationship between at least some of the located features.
9 more elements
An apparatus useful in robotics, the apparatus comprising:
- a single camera operable to capture a number of images of a calibration object;
- means for calibrating the camera, by:
- determining a set of intrinsic parameters of the camera from at least one of the number of images of the calibration object captured by the camera;
- and
- determining a set of extrinsic parameters of the camera from at least one of the number of images of the calibration object captured by the camera, the set of extrinsic parameters comprising a camera space-to-training space transformation defining a transformation between a camera space reference frame and a training space reference frame;
- and
- means for estimating a pose of a target object, by:
- capturing a two-dimensional image of a volume containing a target object;
- locating at least five features in the captured image of the target object;
- and
- determining an object space-to-camera space transformation based at least in part on a position of at least some of the located features using the captured image without any additional captured images and an algorithm that employs a known or determinable physical relationship between at least some of the located features.
8 more elements
Dependencies are inferred from the claim text.