Object pickup strategies for a robotic device

• US 9,333,649 B1
• Filed: 03/14/2014
• Issued: 05/10/2016
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Intrinsic Innovation LLC (Alphabet Inc.)
• Original Assignee: Industrial Perception, Inc. (Alphabet Inc.)
• Inventors: Croft, Steve, Zevenbergen, John, Bradski, Gary, Konolige, Kurt, Rublee, Ethan, Straszheim, Troy, Strasdat, Hauke, Hinterstoisser, Stefan
• Primary Examiner(s): Tran, Khoi
• Assistant Examiner(s): Peche, Jorge
• Application Number:
  US14/213,089
• Time in Patent Office: 788 Days
• Field of Search: 700/214, 700/218, 700/245, 700/253, 700/255, 700/259, 901/40, 901/47
• US Class Current: 1/1
• CPC Class Codes: B25J 19/00 Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators safety-devices in general F16P; protection against radiation in general G21F
• CPC Class Codes: B25J 19/021 Optical sensing devices
• CPC Class Codes: B25J 5/00 Manipulators mounted on wheels or on carriages B25J1/00 takes precedence; programme-controlled manipulators B25J9/00 ; vehicle aspects B60, B62, e.g. remote-controlled steering for motor vehicles B62D1/24; control of position of vehicles G05D1/00
• CPC Class Codes: B25J 9/0093 co-operating with conveyor means
• CPC Class Codes: B25J 9/1612 characterised by the hand, wrist, grip control
• CPC Class Codes: B25J 9/162 Mobile manipulator, movable base with manipulator arm mounted on it
• CPC Class Codes: B25J 9/163 learning, adaptive, model based, rule based expert control
• CPC Class Codes: B25J 9/1664 characterised by motion, path, trajectory planning
• CPC Class Codes: B25J 9/1671 characterised by simulation, either to verify existing program or to create and verify new program, CAD/CAM oriented, graphic oriented programming systems
• CPC Class Codes: B25J 9/1687 Assembly, peg and hole, palletising, straight line, weaving pattern movement
• CPC Class Codes: B25J 9/1694 characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
• CPC Class Codes: B25J 9/1697 Vision controlled systems
• CPC Class Codes: B65G 41/008 frames mounted on wheels or caterpillar for movement on rail tracks B65G41/02
• CPC Class Codes: B65G 47/46 and distributing, e.g. automatically, to desired points in tube mail systems B65G51/36
• CPC Class Codes: B65G 47/50 according to destination signals stored in separate systems
• CPC Class Codes: B65G 61/00 Use of pick-up or transfer devices or of manipulators for stacking or de-stacking articles not otherwise provided for
• CPC Class Codes: B65H 67/065 Manipulators with gripping or holding means for transferring the packages from one station to another, e.g. from a conveyor to a creel trolley
• CPC Class Codes: G01B 11/254 Projection of a pattern, viewing through a pattern, e.g. moiré
• CPC Class Codes: G05B 2219/31312 Identify pallet, bag, box code
• CPC Class Codes: G05B 2219/39391 Visual servoing, track end effector with camera image feedback
• CPC Class Codes: G05B 2219/40053 : Pick 3-D object from pile of objects
• CPC Class Codes: G05B 2219/40298 : Manipulator on vehicle, wheels, mobile
• CPC Class Codes: G05B 2219/40442 : Voxel map, 3-D grid map
• CPC Class Codes: G05B 2219/40543 : Identification and location, position of components, objects
• CPC Class Codes: G06T 17/00 : Three dimensional [3D] modelling, e.g. data description of 3D objects
• CPC Class Codes: G06T 19/003 : Navigation within 3D models or images
• CPC Class Codes: G06T 2200/04 : involving 3D image data
• CPC Class Codes: G06T 7/13 : Edge detection
• CPC Class Codes: G06T 7/529 : from texture
• CPC Class Codes: G06T 7/55 : from multiple images
• CPC Class Codes: G06T 7/593 : from stereo images
• CPC Class Codes: G06T 7/60 : Analysis of geometric attributes
• CPC Class Codes: G06V 20/10 : Terrestrial scenes scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59
• CPC Class Codes: G06V 20/64 : Three-dimensional objects
• CPC Class Codes: H04N 13/239 : using two 2D image sensors having a relative position equal to or related to the interocular distance H04N13/243 takes precedence
• CPC Class Codes: H04N 2013/0081 : Depth or disparity estimation from stereoscopic image signals
• CPC Class Codes: H04N 5/33 : Transforming infrared radiation cameras or camera modules for generating image signals from infrared radiation H04N23/20; circuitry of SSIS for transforming infrared radiation into image signals H04N25/20
• CPC Class Codes: Y10S 901/01 : Mobile robot
• CPC Class Codes: Y10S 901/02 : Arm motion controller
• CPC Class Codes: Y10S 901/06 : Communication with another machine
• CPC Class Codes: Y10S 901/09 : Closed loop, sensor feedback controls arm movement
• CPC Class Codes: Y10S 901/47 : Optical

Abstract

Example embodiments may relate to methods and systems for selecting a grasp point on an object. In particular, a robotic manipulator may identify characteristics of a physical object within a physical environment. Based on the identified characteristics, the robotic manipulator may determine potential grasp points on the physical object corresponding to points at which a gripper attached to the robotic manipulator is operable to grip the physical object. Subsequently, the robotic manipulator may determine a motion path for the gripper to follow in order to move the physical object to a drop-off location for the physical object and then select a grasp point, from the potential grasp points, based on the determined motion path. After selecting the grasp point, the robotic manipulator may grip the physical object at the selected grasp point with the gripper and move the physical object through the determined motion path to the drop-off location.

Claims

1. A method comprising:

identifying one or more characteristics of a physical object within a physical environment, wherein the identified one or more characteristics of the physical object comprise a set of geometric characteristics from a particular perspective viewpoint of the physical object;

making a comparison between the set of geometric characteristics and one or more virtual geometric shapes from the particular perspective viewpoint of the physical object;

based on an output of the comparison indicating that at least one of the geometric characteristics substantially matches a given virtual geometric shape, generating a virtual object representative of the physical object and associated with the matching virtual geometric shape;

adjusting one or more identified characteristics of the physical object based on characteristics of the virtual object;

based on the adjusted identified one or more characteristics, determining by a computing device two or more potential grasp points on the physical object corresponding to points at which a gripper attached to a robotic manipulator is operable to grip the physical object;

determining a motion path for the gripper to follow in order to move the physical object to a drop-off location for the physical object;

selecting a grasp point from the two or more potential grasp points based on the determined motion path; and

providing instructions to cause the robotic manipulator to grip the physical object at the selected grasp point with the gripper and move the physical object through the determined motion path to the drop-off location.

2. The method of claim 1, further comprising:

receiving at least one sensor scan of the physical object, wherein identifying the one or more characteristics of the physical object is based at least in part on the at least one sensor scan of the physical object.

3. The method of claim 1, wherein determining the two or more potential grasp points comprises identifying of two or more flat areas on the physical object.

4. The method of claim 3, wherein the gripper uses suction to grip objects, and wherein the identified two or more flat areas are each larger in area than a gripping surface of the gripper.

5. The method of claim 1, wherein determining the two or more potential grasp points comprises identifying a first point on a first surface of the physical object and a second point on a second surface of the physical object, wherein the first and second surfaces have opposite surface orientations, and wherein the gripper comprises a two finger gripper configured to clamp onto the identified first and second points.

6. The method of claim 1, wherein selecting the grasp point from the two or more potential grasp points further comprises rejecting, from among the two or more potential grasp points, one or more potential grasp points that result in a collision with an obstructing object when the robotic manipulator moves towards the physical object in order to grip the physical object at the selected grasp point with the gripper.

7. The method of claim 1, wherein selecting the grasp point from the two or more potential grasp points further comprises rejecting one or more potential grasp points that result in a collision with an obstructing object when the robotic manipulator moves the physical object through the determined motion path to the drop-off location.

8. The method of claim 1, wherein selecting the grasp point from the two or more potential grasp points further comprises selecting the grasp point based on geometric characteristics of the drop-off location to cause the one or more identified characteristics of the physical object to accommodate the geometric characteristics of the drop-off location.

9. The method of claim 1, wherein determining the motion path for the gripper comprises determining a sequence of joint angles for the robotic manipulator that minimizes a cost function, wherein the cost function comprises a plurality of weighted criterion associated with different characteristics of the motion path.

10. The method of claim 1, wherein determining the motion path for the gripper comprises determining a sequence of Cartesian coordinates for the physical object that minimizes a cost function, wherein the cost function comprises a plurality of weighted criterion associated with different characteristics of the motion path.

11. The method of claim 1, wherein selecting the grasp point from the two or more potential grasp points comprises selecting a grasp point from the two or more potential grasp points that minimizes a cost function, wherein the cost function comprises a plurality of weighted criterion associated with different characteristics of the grasp point.

12. The method of claim 1, wherein selecting the grasp point from the two or more potential grasp points comprises selecting a grasp point from the two or more potential grasp points that minimizes a cost function, wherein the cost function comprises:

a plurality of weighted criterion associated with different characteristics of a pose of the robotic manipulator at the grasp point; and

a plurality of weighted criterion associated with different characteristics of a pose of the robotic manipulator at the drop-off location.

13. The method of claim 1, further comprising:

storing information, in a memory storage, comprising one or more of;

(i) the identified one or more characteristics of the physical object, (ii) the adjusted identified one or more characteristics of the physical object, (iii) the selected grasp point, (iv) the determined motion path, and (v) a performance evaluation of the selected grasp point based on the movement of the physical object through the determined motion path to the drop-off location.

14. The method of claim 13, further comprising:

using the stored information to determine other grasp points for the robotic manipulator to use in picking up additional physical objects from the physical environment.

15. The method of claim 13, further comprising:

providing the stored information to a second robotic manipulator to determine grasp points for the second robotic manipulator to use to pick up physical objects from a physical environment.

16. The method of claim 1, further comprising:

determining one or more adjacent physical objects in contact with the physical object; and

providing instructions to cause the robotic manipulator to create space between the physical object and the one or more adjacent physical objects by using the gripper to cause the physical object to alternately move in at least two opposing directions prior to moving the physical object through the determined motion path.

17. The method of claim 1, further comprising:

receiving sensor data from one or more sensors, wherein generating the virtual object is further based at least in part on the sensor data received from one or more sensors.

18. The method of claim 17, wherein the one or more sensors are attached to the robotic manipulator.

19. The method of claim 1, further comprising:

identifying a fiducial mark at a particular point on the virtual object, wherein the fiducial mark indicates a reference point for gripping the physical object; and

determining a point on the physical object corresponding to the fiducial mark as one of the two or more potential grasp points.

20. A system comprising:

a robotic manipulator;

a gripper attached to the robotic manipulator; and

a control system configured to;

identify one or more characteristics of a physical object within a physical environment, wherein the identified one or more characteristics of the physical object comprise a set of geometric characteristics from a particular perspective viewpoint of the physical object;

make a comparison between the set of geometric characteristics and one or more virtual geometric shapes from the particular perspective viewpoint of the physical object;

based on an output of the comparison indicating that at least one of the geometric characteristics substantially matches a given virtual geometric shape, generate a virtual object representative of the physical object and associated with the matching virtual geometric shape;

adjust one or more identified characteristics of the physical object based on characteristics of the virtual object;

based on the adjusted identified one or more characteristics, determine two or more potential grasp points on the physical object corresponding to points at which the gripper is operable to grip the physical object;

determine a motion path for the gripper to follow in order to move the physical object to a drop-off location for the physical object;

select a grasp point from the two or more potential grasp points based on the determined motion path; and

provide instructions to cause the robotic manipulator to grip the physical object at the selected grasp point with the gripper and move the physical object through the determined motion path to the drop-off location.

21. The system of claim 20, wherein the control system is configured to determine the two or more potential grasp points by identifying of two or more flat areas on the physical object.

22. The system of claim 21, wherein the gripper uses suction to grip objects, and wherein the identified two or more flat areas are each larger in area than a gripping surface of the gripper.

23. The system of claim 20, wherein selecting the grasp point from the two or more potential grasp points further comprises selecting the grasp point based on geometric characteristics of the drop-off location to cause the one or more identified characteristics of the physical object to accommodate the geometric characteristics of the drop-off location.

24. A non-transitory computer readable medium having stored therein instructions, that when executed by a computing device, cause the computing device to perform functions comprising:

identifying one or more characteristics of a physical object within a physical environment, wherein the identified one or more characteristics of the physical object comprise a set of geometric characteristics from a particular perspective viewpoint of the physical object;

making a comparison between the set of geometric characteristics and one or more virtual geometric shapes from the particular perspective viewpoint of the physical object;

based on an output of the comparison indicating that at least one of the geometric characteristics substantially matches a given virtual geometric shape, generating a virtual object representative of the physical object and associated with the matching virtual geometric shape;

adjusting one or more identified characteristics of the physical object based on characteristics of the virtual object;

based on the adjusted identified one or more characteristics, determining two or more potential grasp points on the physical object corresponding to points at which a gripper attached to a robotic manipulator is operable to grip the physical object;

determining a motion path for the gripper to follow in order to move the physical object to a drop-off location for the physical object;

selecting a grasp point from the two or more potential grasp points based on the determined motion path; and

providing instructions to cause the robotic manipulator to grip the physical object at the selected grasp point with the gripper and move the physical object through the determined motion path to the drop-off location.