DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . Claims 1-20 as originally filed are pending and have been considered as follows.
Priority
1. This application claims priority from U.S. Provisional Application No. 63/533,378 filed on 08/18/2023 is acknowledge.
Information Disclosure Statement
2. The information disclosure statement (IDS) filed on 12/16/2024 is being considered by the examiner.
Claim Interpretation
3. The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
4. The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
5. This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
“user interaction device” in claims 1, 3, 4, 15, 16, and 17.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
A review of the specification shows the following appears to be the corresponding structure described in the specification for the 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph limitations:
“user interaction device” in claims 1, 3, 4, 15, 16, and 17 corresponds to “user interaction device presents visual and audio feedback to the user and accepts user input and feedback for human-in-the-loop grasp synthesis and execution. User input may include input from a touch display, augmented reality glasses, body-worn sensor networks (for example, and not by way of limitation, augmented reality or virtual reality gloves), speakers/microphones, body-worn cameras, computer mouse and keyboard, joysticks, etc” [0017] and Fig. 1.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
6. The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
7. Claims 8-14 is/are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding claim 8, it is unclear what is meant by the terms “high-level”. It is unclear what constitute as a high-level. For examination purposes, the examiner is interpreting “high-level path planner” as any path planner.
In the art rejection above, the claims have been treated as best understood by the examiner. Any claim not explicitly rejected under this heading is rejected as being dependent on an indefinite claim.
Claim Rejections - 35 USC § 102
8. In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
9. Claims 1-3, and 6-7 is/are rejected under 35 U.S.C. 102(a)(2)/(a)(1) as being anticipated by Barry et al. (US 20220193894, hereinafter Barry).
Regarding claim 1, Barry teaches a system for grasp synthesis of non-occluded and occluded objects with a camera-equipped robot manipulator (see at least Fig. 1A and abstract: “The method includes receiving a selection input indicating a user-selection of a target object represented in an image corresponding to the space. The target object is for grasping by an end-effector of a robotic manipulator of the robot.”; [0024]: “The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”), said system comprising:
a robot manipulator with a gripper and gripper camera (see at least Fig. 1A and [0024]: “In order to maneuver about the environment 10 or to perform tasks using the arm 126, the robot 100 includes a sensor system 130 with one or more sensors 132, 132a-n (e.g., shown as a first sensor 132, 132a and a second sensor 132, 132b). The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”), wherein said robot manipulator is configured to execute a grasp of a target object (see at least [0022]: “In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10.”; [0035]: “In some implementations, the target object selected by the user corresponds to a respective object for an end-effector 150 of a robotic manipulator of the robot 100 to grasp.”);
a user interaction device configured to present visual and audio feedback to a user and accept user input and feedback (see at least Fig. 1B and [0033]: “A user 12 may interact with the robot 100 via the remote controller 20 that communicates with the robot 100 to perform actions. Additionally, the robot 100 may communicate with the remote controller 20 to display an image on a user interface 300 (e.g., UI 300) of the remote controller 20.”; [0034]: “The image displayed on the UI 300 may include one or more objects that are present in the environment 10 (e.g., within a field of view F.sub.V for a sensor 132 of the robot 100)…The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”; [0067]: “To provide for interaction with a user, one or more aspects of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen for displaying information to the user and optionally a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.”);
at least one processor in communication with said robot manipulator and said user interaction device (see at least Figs. 1B, 4, and [0030]: “In some implementations, as shown in FIGS. 1A and 1B, the robot 100 includes a control system 170. The control system 170 may be configured to communicate with systems of the robot 100, such as the at least one sensor system 130. The control system 170 may perform operations and other functions using hardware 140.”; [0033]: “A user 12 may interact with the robot 100 via the remote controller 20 that communicates with the robot 100 to perform actions.”; [0059]: “The computing device 400 includes a processor 410 (e.g., data processing hardware), memory 420 (e.g., memory hardware), a storage device 430, a high-speed interface/controller 440 connecting to the memory 420 and high-speed expansion ports 450, and a low speed interface/controller 460 connecting to a low speed bus 470 and a storage device 430.”); and
at least one memory in communication with said at least one processor, configured to receive and store data from said robot manipulator and said user interaction device (see at least Figs. 1B, 4, and [0059]: “The computing device 400 includes a processor 410 (e.g., data processing hardware), memory 420 (e.g., memory hardware), a storage device 430, a high-speed interface/controller 440 connecting to the memory 420 and high-speed expansion ports 450, and a low speed interface/controller 460 connecting to a low speed bus 470 and a storage device 430….The processor 410 can process instructions for execution within the computing device 400, including instructions stored in the memory 420 or on the storage device 430 to display graphical information for a graphical user interface (GUI) on an external input/output device, such as display 480 coupled to high speed interface 440.”; [0060]: “The memory 420 stores information non-transitorily within the computing device 400. The memory 420 may be a computer-readable medium, a volatile memory unit(s), or non-volatile memory unit(s). The non-transitory memory 420 may be physical devices used to store programs (e.g., sequences of instructions) or data (e.g., program state information) on a temporary or permanent basis for use by the computing device 400.”).
Regarding claim 2, Barry teaches the limitations of claim 1. Barry further teaches wherein said robot manipulator is a robotic arm comprising a plurality of arm base actuators, an arm first link, an arm second link connected to said arm first link via an elbow joint, wrist actuators (see at least Fig. 1A and [0022]: “To illustrate an example, FIG. 1A depicts the arm 126 with three members 128 corresponding to a lower member 128.sub.L, an upper member 128.sub.U, and a hand member 128.sub.H (e.g., shown as an end-effector 150). Here, the lower member 128.sub.L may rotate or pivot about a first arm joint J.sub.A1 located adjacent to the body 110 (e.g., where the arm 126 connects to the body 110 of the robot 100). The lower member 128.sub.L is coupled to the upper member 128.sub.U at a second arm joint J.sub.A2 and the upper member 128.sub.U is coupled to the hand member 128.sub.H at a third arm joint J.sub.A3. In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10….In other words, the fourth joint J.sub.A4 may function as a twist joint similarly to the third joint J.sub.A3 or wrist joint of the arm 126 adjacent the hand member 128.sub.H. For instance, as a twist joint, one member coupled at the joint J may move or rotate relative to another member coupled at the joint J (e.g., a first member coupled at the twist joint is fixed while the second member coupled at the twist joint rotates). In some implementations, the arm 126 connects to the robot 100 at a socket on the body 110 of the robot 100.”), and at least one gripper including gripper jaws, a gripper camera (see at least Fig. 1A and [0022]: “In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10. The moveable jaw is configured to move relative to the fixed jaw in order to move between an open position for the gripper and a closed position for the gripper (e.g., closed around an object).”; [0024]: “In order to maneuver about the environment 10 or to perform tasks using the arm 126, the robot 100 includes a sensor system 130 with one or more sensors 132, 132a-n (e.g., shown as a first sensor 132, 132a and a second sensor 132, 132b). The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”).
Regarding claim 3, Barry teaches the limitations of claim 1. Barry further teaches wherein said user interaction device comprises a touch display (see at least Fig. 1B and [0034]: “The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”; [0053]: “The grasp geometry generator 210 receives the user-selected target object from the UI 300 and sensor data 134 (e.g., three-dimensional point cloud). The user 12 selects the target object on the two-dimensional image on the UI 300 that corresponds to the three-dimensional point cloud of data 134 for the field of view F.sub.V of the robot 100.”; [0067]: “To provide for interaction with a user, one or more aspects of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen for displaying information to the user and optionally a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer.”).
Regarding claim 6, Barry teaches the limitations of claim 1. Barry further teaches wherein the robot manipulator is mounted on a mobile base (see at least Fig. 1A and [0020]: “Referring to FIGS. 1A and 1B, the robot 100 includes a body 110 with locomotion based structures such as legs 120a-d coupled to the body 110 that enable the robot 100 to move about the environment 10. In some examples, each leg 120 is an articulable structure such that one or more joints J permit members 122 of the leg 120 to move.”).
Regarding claim 7, Barry teaches the limitations of claim 6. Barry further teaches wherein said mobile base is a legged robot (see at least Fig. 1A and [0020]: “Referring to FIGS. 1A and 1B, the robot 100 includes a body 110 with locomotion based structures such as legs 120a-d coupled to the body 110 that enable the robot 100 to move about the environment 10. In some examples, each leg 120 is an articulable structure such that one or more joints J permit members 122 of the leg 120 to move.”).
Claim Rejections - 35 USC § 103
10. In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
11. Claim 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) in view of Hoffman et al. (US 20170217021, hereinafter Hoffman).
Regarding claim 4, Barry teaches the limitations of claim 1.
Barry fails to explicitly teach wherein said user interaction device comprises at least one joystick.
However, Hoffman teaches a method and apparatus for remotely operating a mobile robot with a user interaction device that comprises at least one joystick (see at least Figs. 1, 4A, and [0076]: “Referring to FIGS. 4A-4M, the OCU 100 may include a display (e.g., LCD or touch screen) 110, a keyboard, and one or more auxiliary user inputs, such as a joystick or gaming unit in communication with the computing device 102. As shown, the OCU 100 is a touch screen tablet. The OCU 100 provides a user interface of the teleoperation software application 101 that is rendered on the display 110 of the OCU 100 and allows an operator or user 10 to control the robot 200 from a distance.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Hoffman and provide a user interaction device that comprises at least one joystick, with a reasonable expectation of success, in order to provide an alternative means of controlling the robot.
12. Claims 5, 15, 16, and 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) in view of Ku et al. (US 20220016767, hereinafter Ku).
Regarding claim 5, Barry teaches the limitations of claim 1. Barry further teaches wherein said robot manipulator is further configured to clear a plurality of objects (see at least [0022]: “In the examples shown, the robot 100 includes an arm 126 that functions as a robotic manipulator. The arm 126 may be configured to move about multiple degrees of freedom in order to engage elements of the environment 10 (e.g., objects within the environment 10).”).
Barry fails to explicitly teach clearing a plurality of objects occluding said target object.
However, Ku teaches a method and system for object grasping with a robot that comprises a robot manipulator to clear a plurality of objects occluding a target object (see at least Figs. 2, 8, and [0025]: “First, variants of the system and method enable object grasping from a bin of objects, wherein the objects can be overlapping with other objects and in any random pose. Such variants can improve the grasp success rate for highly occluded objects and/or object scenes (e.g., an example is shown in FIG. 8) by avoiding difficult-to-grasp features of an object and/or by avoiding overlapping objects.”; [0029]: “The method is preferably performed using a system, an example of which is shown in FIG. 2, including: an end effector 110, a robot arm 120, a sensor suite 130, a computing system 140, and/or any other suitable components. The system functions to enable selection of a candidate grasp location and/or articulate the robot arm to grasp an object 104 associated with the grasp location.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Ku and provide a means to clear a plurality of objects occluding a target object, with a reasonable expectation of success, in order to make a path for the robot to access the target object.
Regarding claim 15, Barry teaches a system for grasp synthesis of non-occluded and occluded objects with a camera-equipped robot manipulator (see at least Fig. 1A and abstract: “The method includes receiving a selection input indicating a user-selection of a target object represented in an image corresponding to the space. The target object is for grasping by an end-effector of a robotic manipulator of the robot.”; [0024]: “The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”), said system comprising:
a robotic arm comprising a plurality of arm base actuators, an arm first link, an arm second link connected to said arm first link via an elbow joint, wrist actuators (see at least Fig. 1A and [0022]: “To illustrate an example, FIG. 1A depicts the arm 126 with three members 128 corresponding to a lower member 128.sub.L, an upper member 128.sub.U, and a hand member 128.sub.H (e.g., shown as an end-effector 150). Here, the lower member 128.sub.L may rotate or pivot about a first arm joint J.sub.A1 located adjacent to the body 110 (e.g., where the arm 126 connects to the body 110 of the robot 100). The lower member 128.sub.L is coupled to the upper member 128.sub.U at a second arm joint J.sub.A2 and the upper member 128.sub.U is coupled to the hand member 128.sub.H at a third arm joint J.sub.A3. In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10….In other words, the fourth joint J.sub.A4 may function as a twist joint similarly to the third joint J.sub.A3 or wrist joint of the arm 126 adjacent the hand member 128.sub.H. For instance, as a twist joint, one member coupled at the joint J may move or rotate relative to another member coupled at the joint J (e.g., a first member coupled at the twist joint is fixed while the second member coupled at the twist joint rotates). In some implementations, the arm 126 connects to the robot 100 at a socket on the body 110 of the robot 100.”), and at least one gripper including gripper jaws, a gripper camera (see at least Fig. 1A and [0022]: “In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10. The moveable jaw is configured to move relative to the fixed jaw in order to move between an open position for the gripper and a closed position for the gripper (e.g., closed around an object).”; [0024]: “In order to maneuver about the environment 10 or to perform tasks using the arm 126, the robot 100 includes a sensor system 130 with one or more sensors 132, 132a-n (e.g., shown as a first sensor 132, 132a and a second sensor 132, 132b). The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”), wherein said robotic arm is configured to execute a grasp of a target object, and wherein said grasp of said target object is determined via a method for grasp synthesis (see at least [0022]: “In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10.”; [0035]: “In some implementations, the target object selected by the user corresponds to a respective object for an end-effector 150 of a robotic manipulator of the robot 100 to grasp.”), wherein said method comprises:
estimating surface of a target object (see at least [0032]: “Referring now to FIG. 1B, the sensor system 130 of the robot 100 generates a three-dimensional point cloud of sensor data 134 for an area or space or volume within the environment 10 about the robot 100. Although referred to as a three-dimensional point cloud of sensor data 134, it should be understood that the sensor data 134 may represent a three-dimensional portion of the environment 10 or a two-dimensional portion (such as a surface or plane) of the environment 10. In other words, the sensor data 134 may be a three-dimensional point cloud or a two-dimensional collection of points.”);
identifying an appropriate grasp proposal method (see at least Figs. 2A, 2B, and [0040]: “In some implementations, after the robot 100 begins to execute the initial grasp geometry 212I, the grasping system 200 may determine a new grasp geometry 212N to grasp the target object. The grasping system 200 may determine, after the robot 100 begins execution of the initial grasp geometry 212I, a new grasp geometry 212N that improves and/or refines the grasp geometry 212 being executed. Here, an improvement or refinement in the grasp geometry 212 may correspond to a grasp geometry 212 that is more efficient (e.g., a more cost effective grasp in terms of energy or motion), has a higher likelihood of success, has a more optimal execution time (e.g., faster or slower), etc., to grasp the target object when compared to the initial grasp geometry 212I.”) and grasp scoring method (see at least Figs. 2A, 2B, and [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area. In particular, the grasp geometry generator 210 generates multiple candidate grasp geometries 212 and the grasping system 200 determines which of the multiple candidate grasp geometries 212 the robot 100 should use to grasp the target object. In these implementations, the grasping system 200 includes a scorer 240 that assigns a grasping score 242 to each of the plurality of candidate grasp geometries 212. The grasping score 242 indicates an estimated or projected likelihood of success that the candidate grasp geometry 212 will successfully grasp the target object.”);
proposing a series of grasp candidates according to a selected grasp proposal method (see at least Figs. 2A, 2B, and [0040]: “In some implementations, after the robot 100 begins to execute the initial grasp geometry 212I, the grasping system 200 may determine a new grasp geometry 212N to grasp the target object. The grasping system 200 may determine, after the robot 100 begins execution of the initial grasp geometry 212I, a new grasp geometry 212N that improves and/or refines the grasp geometry 212 being executed. Here, an improvement or refinement in the grasp geometry 212 may correspond to a grasp geometry 212 that is more efficient (e.g., a more cost effective grasp in terms of energy or motion), has a higher likelihood of success, has a more optimal execution time (e.g., faster or slower), etc., to grasp the target object when compared to the initial grasp geometry 212I.”; [0044]: “In some examples, the adjuster 230 determines to continue execution of the initial grasp geometry 212I. In other examples, the adjuster 230 determines to modify the initial grasp geometry 212I to generate a modified grasp geometry 212M. That is, after receiving the updated sensor data 134U, the adjuster 230 compares the initial grasp geometry 212I and the new candidate grasp geometry 212N and determines that it should modify the initial grasp geometry 212I.”; [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area.”);
assigning a quality score to each of said grasp candidates proposed according to a selected grasp scoring method (see at least Figs. 2A, 2B, and [0045]: “In particular, the grasp geometry generator 210 generates multiple candidate grasp geometries 212 and the grasping system 200 determines which of the multiple candidate grasp geometries 212 the robot 100 should use to grasp the target object. In these implementations, the grasping system 200 includes a scorer 240 that assigns a grasping score 242 to each of the plurality of candidate grasp geometries 212. The grasping score 242 indicates an estimated or projected likelihood of success that the candidate grasp geometry 212 will successfully grasp the target object.”; [0047]: “The selector 220 is configured to select the respective candidate grasp geometry 212 with a greatest grasping score 242 as a grasp geometry 212 for the robot 100 to use to grasp the target object. The grasping score 242 may be generated by a scoring algorithm that accounts for different factors that identify an overall performance for a given grasping geometry 212…As an example, the selector 220 receives three candidate grasp geometries 212 that include grasping scores 242 of 0.6, 0.4, and 0.8. In this example, the selector 220 determines the candidate grasp geometry 212 with the grasping score 0.8 has the highest likelihood to successfully grasp the target object. The selector 220 sends the selected candidate grasp geometry 212 (e.g., initial grasp geometry 212I) from the plurality of candidate grasp geometries 212 to the control system 170. The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”); and
applying a post-processing and filtering method to said grasp candidates (see at least [0047]: “The selector 220 is configured to select the respective candidate grasp geometry 212 with a greatest grasping score 242 as a grasp geometry 212 for the robot 100 to use to grasp the target object. The grasping score 242 may be generated by a scoring algorithm that accounts for different factors that identify an overall performance for a given grasping geometry 212…As an example, the selector 220 receives three candidate grasp geometries 212 that include grasping scores 242 of 0.6, 0.4, and 0.8. In this example, the selector 220 determines the candidate grasp geometry 212 with the grasping score 0.8 has the highest likelihood to successfully grasp the target object. The selector 220 sends the selected candidate grasp geometry 212 (e.g., initial grasp geometry 212I) from the plurality of candidate grasp geometries 212 to the control system 170. The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”);
a user interaction device configured to present visual and audio feedback to a user and accept user input and feedback (see at least Fig. 1B and [0033]: “A user 12 may interact with the robot 100 via the remote controller 20 that communicates with the robot 100 to perform actions. Additionally, the robot 100 may communicate with the remote controller 20 to display an image on a user interface 300 (e.g., UI 300) of the remote controller 20.”; [0034]: “The image displayed on the UI 300 may include one or more objects that are present in the environment 10 (e.g., within a field of view F.sub.V for a sensor 132 of the robot 100)…The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”; [0067]: “To provide for interaction with a user, one or more aspects of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen for displaying information to the user and optionally a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.”);
at least one processor in communication with said robot manipulator and said user interaction device (see at least Figs. 1B, 4, and [0030]: “In some implementations, as shown in FIGS. 1A and 1B, the robot 100 includes a control system 170. The control system 170 may be configured to communicate with systems of the robot 100, such as the at least one sensor system 130. The control system 170 may perform operations and other functions using hardware 140.”; [0033]: “A user 12 may interact with the robot 100 via the remote controller 20 that communicates with the robot 100 to perform actions.”; [0059]: “The computing device 400 includes a processor 410 (e.g., data processing hardware), memory 420 (e.g., memory hardware), a storage device 430, a high-speed interface/controller 440 connecting to the memory 420 and high-speed expansion ports 450, and a low speed interface/controller 460 connecting to a low speed bus 470 and a storage device 430.”); and
at least one memory in communication with said at least one processor, configured to receive and store data from said robot manipulator and said user interaction device (see at least Figs. 1B, 4, and [0059]: “The computing device 400 includes a processor 410 (e.g., data processing hardware), memory 420 (e.g., memory hardware), a storage device 430, a high-speed interface/controller 440 connecting to the memory 420 and high-speed expansion ports 450, and a low speed interface/controller 460 connecting to a low speed bus 470 and a storage device 430….The processor 410 can process instructions for execution within the computing device 400, including instructions stored in the memory 420 or on the storage device 430 to display graphical information for a graphical user interface (GUI) on an external input/output device, such as display 480 coupled to high speed interface 440.”; [0060]: “The memory 420 stores information non-transitorily within the computing device 400. The memory 420 may be a computer-readable medium, a volatile memory unit(s), or non-volatile memory unit(s). The non-transitory memory 420 may be physical devices used to store programs (e.g., sequences of instructions) or data (e.g., program state information) on a temporary or permanent basis for use by the computing device 400.”).
Barry fails to explicitly teach estimating surface normals of a target object.
However, Ku teaches a method and system for object grasping with a robot that comprises a robot manipulator that estimates surface normals of a target object (see at least [0064]: “In a first variant, S100 can include capturing one or more images of the inference scene and/or retrieving one or more images of the inference scene stored during a previous cycle of the method; optionally determining a 3D point cloud and optionally surface normals for each point of the point cloud based on the one or more images and/or contemporaneously sampled depth information; processing the one or more images, the point cloud, and/or surface normals, using the detector to determine the key points and the occlusion score.”; [0085]: “The graspability score can be determined based on: a grasp probability score, an object (or keypoint) detection score (e.g., detected or not), the occlusion score, the candidate grasp location's corresponding 3D location (e.g., height of the 3D grasp location), proximity to the edge of a scene, proximity to edge of the object, grasp location's associated surface normal, whether the depths within a predetermined radius of the candidate grasp location are within a predetermined range of the grasp location's depth (e.g., the surface planarity), and/or any other suitable parameter.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Ku and provide a means to estimate surface normals of a target object, with a reasonable expectation of success, in order to consider the surface information of the object when assigning a grasp score [0085].
Regarding claim 16, modified Barry teaches the limitations of claim 15. Barry further teaches wherein said user interaction device comprises a touch display (see at least Fig. 1B and [0034]: “The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”; [0053]: “The grasp geometry generator 210 receives the user-selected target object from the UI 300 and sensor data 134 (e.g., three-dimensional point cloud). The user 12 selects the target object on the two-dimensional image on the UI 300 that corresponds to the three-dimensional point cloud of data 134 for the field of view F.sub.V of the robot 100.”; [0067]: “To provide for interaction with a user, one or more aspects of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display) monitor, or touch screen for displaying information to the user and optionally a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer.”).
Regarding claim 18, modified Barry teaches the limitations of claim 15. Barry further teaches wherein said robot manipulator is further configured to clear a plurality of objects (see at least [0022]: “In the examples shown, the robot 100 includes an arm 126 that functions as a robotic manipulator. The arm 126 may be configured to move about multiple degrees of freedom in order to engage elements of the environment 10 (e.g., objects within the environment 10).”).
Barry fails to explicitly teach clearing a plurality of objects occluding said target object.
However, Ku teaches a method and system for object grasping with a robot that comprises a robot manipulator to clear a plurality of objects occluding a target object (see at least Figs. 2, 8, and [0025]: “First, variants of the system and method enable object grasping from a bin of objects, wherein the objects can be overlapping with other objects and in any random pose. Such variants can improve the grasp success rate for highly occluded objects and/or object scenes (e.g., an example is shown in FIG. 8) by avoiding difficult-to-grasp features of an object and/or by avoiding overlapping objects.”; [0029]: “The method is preferably performed using a system, an example of which is shown in FIG. 2, including: an end effector 110, a robot arm 120, a sensor suite 130, a computing system 140, and/or any other suitable components. The system functions to enable selection of a candidate grasp location and/or articulate the robot arm to grasp an object 104 associated with the grasp location.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Ku and provide a means to clear a plurality of objects occluding a target object, with a reasonable expectation of success, in order to make a path for the robot to access the target object.
Regarding claim 19, modified Barry teaches the limitations of claim 15. Barry further teaches wherein the robot manipulator is mounted on a mobile base (see at least Fig. 1A and [0020]: “Referring to FIGS. 1A and 1B, the robot 100 includes a body 110 with locomotion based structures such as legs 120a-d coupled to the body 110 that enable the robot 100 to move about the environment 10. In some examples, each leg 120 is an articulable structure such that one or more joints J permit members 122 of the leg 120 to move.”).
Regarding claim 20, modified Barry teaches the limitations of claim 19. Barry further teaches wherein said mobile base is a legged robot (see at least Fig. 1A and [0020]: “Referring to FIGS. 1A and 1B, the robot 100 includes a body 110 with locomotion based structures such as legs 120a-d coupled to the body 110 that enable the robot 100 to move about the environment 10. In some examples, each leg 120 is an articulable structure such that one or more joints J permit members 122 of the leg 120 to move.”).
13. Claims 8 and 13-14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) and Ku et al. (US 20220016767, hereinafter Ku) in further view of Prats (US 20170326728, hereinafter Prats).
Regarding claim 8, Barry teaches a method for grasp synthesis of non-occluded and occluded objects with a camera-equipped robot manipulator (see at least Fig. 1A and abstract: “The method includes receiving a selection input indicating a user-selection of a target object represented in an image corresponding to the space. The target object is for grasping by an end-effector of a robotic manipulator of the robot.”; [0024]: “The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”), said method comprising:
estimating surface of a target object (see at least [0032]: “Referring now to FIG. 1B, the sensor system 130 of the robot 100 generates a three-dimensional point cloud of sensor data 134 for an area or space or volume within the environment 10 about the robot 100. Although referred to as a three-dimensional point cloud of sensor data 134, it should be understood that the sensor data 134 may represent a three-dimensional portion of the environment 10 or a two-dimensional portion (such as a surface or plane) of the environment 10. In other words, the sensor data 134 may be a three-dimensional point cloud or a two-dimensional collection of points.”);
identifying an appropriate grasp proposal method (see at least Figs. 2A, 2B, and [0040]: “In some implementations, after the robot 100 begins to execute the initial grasp geometry 212I, the grasping system 200 may determine a new grasp geometry 212N to grasp the target object. The grasping system 200 may determine, after the robot 100 begins execution of the initial grasp geometry 212I, a new grasp geometry 212N that improves and/or refines the grasp geometry 212 being executed. Here, an improvement or refinement in the grasp geometry 212 may correspond to a grasp geometry 212 that is more efficient (e.g., a more cost effective grasp in terms of energy or motion), has a higher likelihood of success, has a more optimal execution time (e.g., faster or slower), etc., to grasp the target object when compared to the initial grasp geometry 212I.”) and grasp scoring method (see at least Figs. 2A, 2B, and [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area. In particular, the grasp geometry generator 210 generates multiple candidate grasp geometries 212 and the grasping system 200 determines which of the multiple candidate grasp geometries 212 the robot 100 should use to grasp the target object. In these implementations, the grasping system 200 includes a scorer 240 that assigns a grasping score 242 to each of the plurality of candidate grasp geometries 212. The grasping score 242 indicates an estimated or projected likelihood of success that the candidate grasp geometry 212 will successfully grasp the target object.”);
proposing a series of grasp candidates according to a selected grasp proposal method (see at least Figs. 2A, 2B, and [0040]: “In some implementations, after the robot 100 begins to execute the initial grasp geometry 212I, the grasping system 200 may determine a new grasp geometry 212N to grasp the target object. The grasping system 200 may determine, after the robot 100 begins execution of the initial grasp geometry 212I, a new grasp geometry 212N that improves and/or refines the grasp geometry 212 being executed. Here, an improvement or refinement in the grasp geometry 212 may correspond to a grasp geometry 212 that is more efficient (e.g., a more cost effective grasp in terms of energy or motion), has a higher likelihood of success, has a more optimal execution time (e.g., faster or slower), etc., to grasp the target object when compared to the initial grasp geometry 212I.”; [0044]: “In some examples, the adjuster 230 determines to continue execution of the initial grasp geometry 212I. In other examples, the adjuster 230 determines to modify the initial grasp geometry 212I to generate a modified grasp geometry 212M. That is, after receiving the updated sensor data 134U, the adjuster 230 compares the initial grasp geometry 212I and the new candidate grasp geometry 212N and determines that it should modify the initial grasp geometry 212I.”; [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area.”);
assigning a quality score to each of said grasp candidates proposed according to a selected grasp scoring method (see at least Figs. 2A, 2B, and [0045]: “In particular, the grasp geometry generator 210 generates multiple candidate grasp geometries 212 and the grasping system 200 determines which of the multiple candidate grasp geometries 212 the robot 100 should use to grasp the target object. In these implementations, the grasping system 200 includes a scorer 240 that assigns a grasping score 242 to each of the plurality of candidate grasp geometries 212. The grasping score 242 indicates an estimated or projected likelihood of success that the candidate grasp geometry 212 will successfully grasp the target object.”; [0047]: “The selector 220 is configured to select the respective candidate grasp geometry 212 with a greatest grasping score 242 as a grasp geometry 212 for the robot 100 to use to grasp the target object. The grasping score 242 may be generated by a scoring algorithm that accounts for different factors that identify an overall performance for a given grasping geometry 212…As an example, the selector 220 receives three candidate grasp geometries 212 that include grasping scores 242 of 0.6, 0.4, and 0.8. In this example, the selector 220 determines the candidate grasp geometry 212 with the grasping score 0.8 has the highest likelihood to successfully grasp the target object. The selector 220 sends the selected candidate grasp geometry 212 (e.g., initial grasp geometry 212I) from the plurality of candidate grasp geometries 212 to the control system 170. The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”);
applying a post-processing and filtering method to said grasp candidates (see at least [0047]: “The selector 220 is configured to select the respective candidate grasp geometry 212 with a greatest grasping score 242 as a grasp geometry 212 for the robot 100 to use to grasp the target object. The grasping score 242 may be generated by a scoring algorithm that accounts for different factors that identify an overall performance for a given grasping geometry 212…As an example, the selector 220 receives three candidate grasp geometries 212 that include grasping scores 242 of 0.6, 0.4, and 0.8. In this example, the selector 220 determines the candidate grasp geometry 212 with the grasping score 0.8 has the highest likelihood to successfully grasp the target object. The selector 220 sends the selected candidate grasp geometry 212 (e.g., initial grasp geometry 212I) from the plurality of candidate grasp geometries 212 to the control system 170. The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”), said post-processing and filtering method comprising:
confirming, via a high-level path planner, that a pre-grasp position to a grasp position is kinematically possible (see at least [0045]: “The generator 210 may generate a plurality of grasp geometries 212 because there are a number of pose permutations possible that enable the end-effector 150 to grasp some portion of the target object. For instance, the end-effector 150 may approach and/or grasp the target object from a particular direction or movement vector in 3D space or at a particular orientation (e.g., pitch, roll, or yaw). In other words, since the end-effector 150 may have multiple degrees of freedom at its disposal to affect the manner in which the end-effector 150 grasps the target object, the generator 210 may generate some number of these permutations as candidate grasp geometries 212.”; [0048]: “The grasping system 200 sends the initial grasp geometry 212I to the control system 170 to initiate a sequence of movements to grasp the target object according to the initial grasp geometry 212I. In other words, to execute the initial grasp geometry 212I, the control system 170 instructs the arm 126 to move from an initial pose of the arm 126 to a grasping pose designated by the initial grasping geometry 212I. Here, the initial pose of the arm 126 refers to the pose or state of the arm 126 when the controller 20 received the input from the user 12 selecting the target object to be grasped by the end-effector 150 of the arm 126.”; [0049]: “In some implementations, the grasping system 200 determines a plurality of new candidate grasp geometries 212N after the robot 100 begins execution on the initial grasp geometry 212I. That is, while the end-effector 150 of the robotic manipulator moves to grasp the target object based on the initial grasp geometry 212I the sensor system 130 receives updated sensor data 134U for a second pose of the end-effector 150 of the robotic manipulator...The grasp geometry generator 210 sends each new candidate grasp geometry 212N to the scorer 240.”); and
removing similar grasps with lower quality scores (see at least [0047]: “As an example, the selector 220 receives three candidate grasp geometries 212 that include grasping scores 242 of 0.6, 0.4, and 0.8. In this example, the selector 220 determines the candidate grasp geometry 212 with the grasping score 0.8 has the highest likelihood to successfully grasp the target object. The selector 220 sends the selected candidate grasp geometry 212 (e.g., initial grasp geometry 212I) from the plurality of candidate grasp geometries 212 to the control system 170. The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”).
Barry fails to explicitly teach estimating surface normals of a target object.
However, Ku teaches a method and system for object grasping with a robot that comprises a robot manipulator that estimates surface normals of a target object (see at least [0064]: “In a first variant, S100 can include capturing one or more images of the inference scene and/or retrieving one or more images of the inference scene stored during a previous cycle of the method; optionally determining a 3D point cloud and optionally surface normals for each point of the point cloud based on the one or more images and/or contemporaneously sampled depth information; processing the one or more images, the point cloud, and/or surface normals, using the detector to determine the key points and the occlusion score.”; [0085]: “The graspability score can be determined based on: a grasp probability score, an object (or keypoint) detection score (e.g., detected or not), the occlusion score, the candidate grasp location's corresponding 3D location (e.g., height of the 3D grasp location), proximity to the edge of a scene, proximity to edge of the object, grasp location's associated surface normal, whether the depths within a predetermined radius of the candidate grasp location are within a predetermined range of the grasp location's depth (e.g., the surface planarity), and/or any other suitable parameter.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Ku and provide a means to estimate surface normals of a target object, with a reasonable expectation of success, in order to consider the surface information of the object when assigning a grasp score [0085].
The combination of Barry and Ku fails to explicitly teach confirming that a path along an approach axis from a pre-grasp position to a grasp position is collision-free and kinematically possible.
However, Prats teaches a method and system for generating a grasp pose for grasping an object by a robot that confirms that a path along an approach axis from a pre-grasp position to a grasp position is collision-free and kinematically possible (see at least [0058]: “The grasp generation engine 118 generates one or more candidate grasp poses of a grasping end effector based on the grasp approach vector and/or other constraints determined by engine 116. For example, the grasp generation engine 118 may employ a collision checker to generate multiple candidate grasp poses that each conform to the grasp approach vector (e.g., with a rotational axis of the grasping end effector aligned with the approach vector) and that do not collide with the object to be grasped and/or with other object(s) in the environment with the object to be grasped. The grasp generation engine 118 may optionally utilize a model of the grasping end effector and/or of other components of the robot to determine conformance to a grasp approach vector and may utilize the model(s) and the 3D point cloud to determine whether the grasping end effector and/or other components of the robot collide with object(s) in the environment.”; [0063]: “FIG. 3 illustrates the robot 180 with the grasping end effector 185 in a grasp pose determined based on disclosed implementations. From the grasp pose, the robot 180 may attempt a grasp of the object by further adjustment of the pose of the end effector 185. For instance, the robot 180 may move the end effector 185 along a path that conforms to the grasp approach vector determined by grasp constraint engine 116 and/or may move actuable members 186A and 186B toward one another to attempt a grasp.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry and Ku to incorporate the teachings of Prats and provide a means to confirm that a path along an approach axis from a pre-grasp position to a grasp position is collision-free and kinematically possible, with a reasonable expectation of success, in order ensure that the appropriate axis is selected that do not collide with the object to be grasped and/or with other object(s) in the environment with the object to be grasped [0058].
Regarding claim 13, modified Barry teaches the limitations of claim 8. Barry further teaches wherein said method further comprises executing a proposed grasp of said target object with an optimal grasp score (see at least [0047]: “The control system 170 instructs the robot 100 to execute the candidate grasp geometry 212 with the grasping score 242 of 0.8 as initial grasp geometry 212I.”; [0052]: “In some implementations, the adjuster 230 only modifies the initial grasp geometry 212I when the grasping score 242 of the new candidate grasp geometry 212N exceeds the score 242 of the initial grasp geometry 212I by a threshold.”; [0057]: “The method 500, at operation 508, includes determining a grasp geometry 212 for the robotic manipulator to grasp the target object within the grasp area 216. The method 500, at operation 510, includes instructing the end-effector 150 of the robotic manipulator to grasp the target object within the grasp area 216 based on the grasp geometry 212.”).
Regarding claim 14, modified Barry teaches the limitations of claim 8. Barry further teaches wherein robot manipulator is a robotic arm comprising a plurality of arm base actuators, an arm first link, an arm second link connected to said arm first link via an elbow joint, wrist actuators (see at least Fig. 1A and [0022]: “To illustrate an example, FIG. 1A depicts the arm 126 with three members 128 corresponding to a lower member 128.sub.L, an upper member 128.sub.U, and a hand member 128.sub.H (e.g., shown as an end-effector 150). Here, the lower member 128.sub.L may rotate or pivot about a first arm joint J.sub.A1 located adjacent to the body 110 (e.g., where the arm 126 connects to the body 110 of the robot 100). The lower member 128.sub.L is coupled to the upper member 128.sub.U at a second arm joint J.sub.A2 and the upper member 128.sub.U is coupled to the hand member 128.sub.H at a third arm joint J.sub.A3. In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10….In other words, the fourth joint J.sub.A4 may function as a twist joint similarly to the third joint J.sub.A3 or wrist joint of the arm 126 adjacent the hand member 128.sub.H. For instance, as a twist joint, one member coupled at the joint J may move or rotate relative to another member coupled at the joint J (e.g., a first member coupled at the twist joint is fixed while the second member coupled at the twist joint rotates). In some implementations, the arm 126 connects to the robot 100 at a socket on the body 110 of the robot 100.”), and at least one gripper including gripper jaws, a gripper camera (see at least Fig. 1A and [0022]: “In some examples, such as FIG. 1A, the hand member 128.sub.H or end-effector 150 is a mechanical gripper that includes a moveable jaw and a fixed jaw configured to perform different types of grasping of elements within the environment 10. The moveable jaw is configured to move relative to the fixed jaw in order to move between an open position for the gripper and a closed position for the gripper (e.g., closed around an object).”; [0024]: “In order to maneuver about the environment 10 or to perform tasks using the arm 126, the robot 100 includes a sensor system 130 with one or more sensors 132, 132a-n (e.g., shown as a first sensor 132, 132a and a second sensor 132, 132b). The sensors 132 may include vision/image sensors, inertial sensors (e.g., an inertial measurement unit (IMU)), force sensors, and/or kinematic sensors. Some examples of sensors 132 include a camera such as a stereo camera, a time-of-flight (TOF) sensor, a scanning light-detection and ranging (LIDAR) sensor, or a scanning laser-detection and ranging (LADAR) sensor.”).
14. Claims 9 and 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) and Ku et al. (US 20220016767, hereinafter Ku) and Prats (US 20170326728, hereinafter Prats) in further view of Sundermeyer et al. (US 20220288783, hereinafter Sundermeyer).
Regarding claim 9, modified Barry teaches the limitations of claim 8. Barry further teaches wherein said grasp proposal method comprises:
computing a point cloud of said target object (see at least [0025]: “The one or more sensors 132 may capture sensor data 134 that defines the three-dimensional point cloud for the area within the environment 10 about the robot”; [0035]: “In some implementations, the target object selected by the user corresponds to a respective object for an end-effector 150 of a robotic manipulator of the robot 100 to grasp. For example, the sensor system 130 of a robot 100 in a manufacturing environment 10 generates a three-dimensional point cloud of sensor data 134 for an area within the manufacturing environment 10. The UI 300 displays the two-dimensional image that corresponds to the three-dimensional point cloud of sensor data 134 within the manufacturing environment 10. The user 12 may instruct the robot 100 to grasp a target object (e.g., a valve) within the manufacturing environment 10 by selecting the target object (e.g., valve) on the UI 300. The remote controller 20 sends the selected target object to the robot 100 to execute the grasp on the target object.”);
generating a gripper pose that places a grasp point at said point cloud (see at least [0053]: “Referring now to FIG. 2C, in some examples, the grasp geometry generator 210 generates the grasp area 216. By generating the grasp area 216, the grasp geometry generator 210 translates the user selected two-dimensional area of interest (e.g., selected target object) into the grasp area 216 in the three-dimensional point cloud of sensor data 134. Specifically, the generation of the grasp area 216 allows the user 12 to interact with the two-dimensional image to instruct the robot 100 to perform an action in the three-dimensional environment 10. The grasp geometry generator 210 receives the user-selected target object from the UI 300 and sensor data 134 (e.g., three-dimensional point cloud). The user 12 selects the target object on the two-dimensional image on the UI 300 that corresponds to the three-dimensional point cloud of data 134 for the field of view F.sub.V of the robot 100. The grasp geometry generator 210 projects a plurality of rays from the selected target object from the two-dimensional image onto the three-dimensional point cloud of sensor data 134. The grasp area 216 therefore corresponds to the area formed by the intersection of the projected rays and the three-dimensional point cloud of sensor data 134”);
generating a plurality of primary poses via translating said gripper pose in cartesian coordinates in random directions (see at least [0036]: “For instance, the grasping system 200 generates a grasp area or grasp region corresponding to the area within the three-dimensional environment 10 where the target object is actually located in to order designate where the end-effector 150 is to grasp the target object…The grasp geometry 212 indicates the pose of the end-effector 150 of the robotic manipulator, where the pose represents the translation (e.g., x-coordinate, y-coordinate, and z-coordinate) and orientation (e.g., pitch, yaw, and roll) of the end-effector 150. That is, the grasp geometry 212 indicates the pose (e.g., orientation and translation) that the end-effector 150 of the robotic manipulator uses to grasp the target object.”; [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area…For instance, the end-effector 150 may approach and/or grasp the target object from a particular direction or movement vector in 3D space or at a particular orientation (e.g., pitch, roll, or yaw). In other words, since the end-effector 150 may have multiple degrees of freedom at its disposal to affect the manner in which the end-effector 150 grasps the target object, the generator 210 may generate some number of these permutations as candidate grasp geometries 212.”);
sampling a plurality of secondary poses around each of said primary poses via rotating said gripper in increments (see at least [0045]: “Referring now to FIG. 2B, in some implementations, the grasp geometry generator 210 generates a plurality of candidate grasp geometries 212, 212a-n based on the selected target object within the grasp area…For instance, the end-effector 150 may approach and/or grasp the target object from a particular direction or movement vector in 3D space or at a particular orientation (e.g., pitch, roll, or yaw). In other words, since the end-effector 150 may have multiple degrees of freedom at its disposal to affect the manner in which the end-effector 150 grasps the target object, the generator 210 may generate some number of these permutations as candidate grasp geometries 212.”); and
repeating said sampling until a desired number of poses for grasp candidates is generated (see at least [0051]: “In some implementations, when the corresponding grasping score 242 of the new candidate grasp geometries 212N exceeds the grasping score 242 of the initial grasp geometry 212I, the adjuster 230 modifies the initial grasp geometry 212I based on the respective candidate grasp geometry 212N from the new set of candidate grasp geometries 212N. For example, the robot 100 begins execution of the initial grasp geometry 212I with a grasping score of 0.8. After the robot 100 begins execution of the initial grasp geometry 212I, the grasp geometry generator 210 receives updated sensor data 134U that corresponds to the current field of view F.sub.V of the one or more sensors 132. The grasp geometry generator 210 generates a plurality of new candidate grasp geometries 212N based on the updated sensor data 134. In this example, the adjuster 230 receives the initial grasp geometry 212I with the grasping score 242 of 0.8 and receives a new candidate grasp geometry 212N with a grasping score 242 of 0.85. Here, the adjuster 230 determines the grasping score 242 (e.g., grasping score 242 of 0.85) for the new candidate grasp geometry 212N exceeds the grasping score 242 (e.g., grasping score 242 of 0.8) of the initial grasp geometry 212I and modifies the initial grasp geometry 212I. As state previously, this modification may make some form of adjustment to the initial grasp geometry 212I or complete replacement of the initial grasp geometry 212I with the new candidate grasp geometry 212N.”).
Barry fails to explicitly teach that the point cloud is a point cloud centroid.
However, Sundermeyer teaches an apparatus and system for machine learning of robotic grasp poses in a cluttered environment that computes a point cloud centroid (see at least [0095]: “In at least one embodiment, local regions of interest can be optionally extracted around the 3D centroid of point cloud segments in order to maximize the number of potential contact points. In at least one embodiment, tubes are extracted with an edge length set to twice the largest spanning dimension, but at least 0.3 m and at most 0.6 m.”; [0089]: “In at least one embodiment, for inference the point cloud is centered at its mean in camera coordinates. In at least one embodiment, training generates 10000 table top scenes by placing 8-12 grasp annotated ShapeNet models at random stable poses. At least one embodiment uses rejection sampling to avoid collisions. At least one embodiment trains with a batch size of 3 for 144.000 iterations which takes ˜40 hours on a single Nvidia V100 GPU. In at least one embodiment, convergence is significantly faster than on previous methods which take up to one week on a single GPU for training.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Sundermeyer and computes a point cloud centroid, with a reasonable expectation of success, in order to maximize the number of potential contact points [0095].
Regarding claim 12, modified Barry teaches the limitations of claim 8. Barry further teaches wherein said grasp proposal method comprises:
using a machine learning algorithm exposed to said target object (see at least [0034]: “In some implementations, the image is classified by a machine learning algorithm in order to identify the presence of one or more graspable objects in the image that correspond to one or more graspable objects within a portion of the environment 10 corresponding to the image. In particular, the sensor system 130 receives the image that corresponds to the area (e.g., environment 10) and sends the image (e.g., sensor data 134) to the grasping system 200. The grasping system 200 classifies graspable objects within the received image (e.g., sensor data 134) using a machine learning object classification algorithm. For example, the grasping system 200 may classify a piece of clothing on the ground as “laundry,” or a piece of trash on the ground as “trash.” The classification of the objects in the image may display to the user 12 on the UI 300.”);
mapping a point cloud to a set of grasp poses (see at least [0036]: “In particular, the grasping system 200 transforms the user-selected target object from the two-dimensional image to the grasp area on the three-dimensional point cloud of sensor data 134. By generating the grasp area, the grasping system 200 allows the selected target object from the two-dimensional image to instruct the robot 100 to grasp the target object in the three-dimensional environment 10. In some configurations, the grasping system 200 generates the grasp area by projecting a plurality of rays from the selected target object of the image onto the three-dimensional point cloud of sensor data 134, as discussed in more detail below in FIG. 2C. After determining the grasp area, the grasping system 200 determines a grasp geometry 212 for the robotic manipulator (i.e., the arm 126 of the robot 100) to grasp the target object with. The grasp geometry 212 indicates the pose of the end-effector 150 of the robotic manipulator, where the pose represents the translation (e.g., x-coordinate, y-coordinate, and z-coordinate) and orientation (e.g., pitch, yaw, and roll) of the end-effector 150. That is, the grasp geometry 212 indicates the pose (e.g., orientation and translation) that the end-effector 150 of the robotic manipulator uses to grasp the target object.”); and
inferring suitable grasp candidates from said point cloud of said target object (see at least [0052]: “In some implementations, the adjuster 230 only modifies the initial grasp geometry 212I when the grasping score 242 of the new candidate grasp geometry 212N exceeds the score 242 of the initial grasp geometry 212I by a threshold. For example, the adjuster 230 only modifies the initial grasp geometry 212I when the grasping score 242 of the new candidate grasp geometry 212N exceeds the grasping score 242 of the initial grasp geometry 212I by a margin of 0.1.”; [0057]: “The method 500, at operation 506, includes generating a grasp area 216 for the end-effector 150 of the robotic manipulator by projecting a plurality of rays 218 from the selected target object of the image onto the three-dimensional point cloud of sensor data 134. The method 500, at operation 508, includes determining a grasp geometry 212 for the robotic manipulator to grasp the target object within the grasp area 216. The method 500, at operation 510, includes instructing the end-effector 150 of the robotic manipulator to grasp the target object within the grasp area 216 based on the grasp geometry 212.”).
Barry fails to explicitly teach training a deep neural network prior to exposure to said target object.
However, Sundermeyer teaches an apparatus and system for machine learning of robotic grasp poses in a cluttered environment that trains a deep neural network prior to exposure to said target object (see at least [0108]: “FIG. 9 illustrates an example of a process that, as a result of being performed by a computer system, trains a machine learned model to grasp an object, in at least one embodiment. In at least one embodiment, at block 902, the computer system obtains a collection of models for different types of objects to be used to train a neural network. In at least one embodiment, the models are 3-D solid object models for objects of various types and sizes similar to those, but not necessarily identical, for which grasps are to be generated. In at least one embodiment, at block 904, the computer system generates a simulation that includes a collection of objects selected from the set of object models. In at least one embodiment, the simulation includes a variety of object types placed randomly in a work area that includes one object to be grasped. In at least one embodiment, at block 906, the computer system examines the objects models directly and generates a plurality of grasp poses for each object in the scene. In at least one embodiment, grasp poses are generated as two point grasps with a given width and gripper orientation.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Sundermeyer and trains a deep neural network prior to exposure to said target object, with a reasonable expectation of success, in order to build a simulation to include a variety of object types for the robot system to recognize and learn.
15. Claim 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) and Ku et al. (US 20220016767, hereinafter Ku) and Prats (US 20170326728, hereinafter Prats) in view of Douillard et al. (US 20200193606, hereinafter Douillard).
Regarding claim 10, modified Barry teaches the limitations of claim 8. Barry further teaches wherein said grasp proposal method comprises:
obtaining an image of said target object (see at least [0034]: “The image displayed on the UI 300 may include one or more objects that are present in the environment 10 (e.g., within a field of view F.sub.V for a sensor 132 of the robot 100)…The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”;
running an analysis on a plurality of pixels of said image (see at least [0054]: “In particular, the grasp geometry generator 210 projects the plurality of rays from one or more pixels of the selected target object. Each ray of the plurality of rays projected from the two-dimensional image to the three-dimensional point cloud represents a pixel of the selected target object. The collection of the plurality of rays in the three-dimensional point cloud represents the grasp area 216. By projecting a ray for each pixel from the selected target object, the grasp geometry generator 210 translates the two-dimensional area of interest for the user 12 (e.g., selected target object) to the three-dimensional grasp area 216. Stated differently, the grasp area 216 designates a three-dimensional area that includes the target object such that the grasping system 200 may generate a grasp geometry 212 to grasp the three-dimensional target object within the grasp area 216. This means that the grasp area 216 designates an area of interest for the robotic manipulator to grasp. From this identified grasp area 216, the grasp geometry generator 210 may use the sensor data 134 within the boundaries of the identified grasp area 216 to understand the target object (e.g., the contour of the target object represented by the 3D point cloud sensor data 134) and to determine the grasp geometry 212.”);
identifying a primary axis and a secondary axis (see at least [0045]: “The generator 210 may generate a plurality of grasp geometries 212 because there are a number of pose permutations possible that enable the end-effector 150 to grasp some portion of the target object. For instance, the end-effector 150 may approach and/or grasp the target object from a particular direction or movement vector in 3D space or at a particular orientation (e.g., pitch, roll, or yaw). In other words, since the end-effector 150 may have multiple degrees of freedom at its disposal to affect the manner in which the end-effector 150 grasps the target object, the generator 210 may generate some number of these permutations as candidate grasp geometries 212.”);
orienting a grasp pose candidate such that a grasp axis aligns with an axis (see at least [0045]: “The generator 210 may generate a plurality of grasp geometries 212 because there are a number of pose permutations possible that enable the end-effector 150 to grasp some portion of the target object. For instance, the end-effector 150 may approach and/or grasp the target object from a particular direction or movement vector in 3D space or at a particular orientation (e.g., pitch, roll, or yaw). In other words, since the end-effector 150 may have multiple degrees of freedom at its disposal to affect the manner in which the end-effector 150 grasps the target object, the generator 210 may generate some number of these permutations as candidate grasp geometries 212.”; [0048]: “The grasping system 200 sends the initial grasp geometry 212I to the control system 170 to initiate a sequence of movements to grasp the target object according to the initial grasp geometry 212I. In other words, to execute the initial grasp geometry 212I, the control system 170 instructs the arm 126 to move from an initial pose of the arm 126 to a grasping pose designated by the initial grasping geometry 212I. Here, the initial pose of the arm 126 refers to the pose or state of the arm 126 when the controller 20 received the input from the user 12 selecting the target object to be grasped by the end-effector 150 of the arm 126.”; [0049]: “In some implementations, the grasping system 200 determines a plurality of new candidate grasp geometries 212N after the robot 100 begins execution on the initial grasp geometry 212I. That is, while the end-effector 150 of the robotic manipulator moves to grasp the target object based on the initial grasp geometry 212I the sensor system 130 receives updated sensor data 134U for a second pose of the end-effector 150 of the robotic manipulator...The grasp geometry generator 210 sends each new candidate grasp geometry 212N to the scorer 240.”).
Barry fails to explicitly teach obtaining an image segmentation mask of said target object; running a principal component analysis on a plurality of pixels of said segmentation mask; and identifying a primary axis explaining maximum variance and a secondary axis along which residuals are minimalized.
However, Douillard teaches a system and method for image analysis that obtains an image segmentation mask of said target object (see at least [0031]: “In some instances, the converting operations described herein may reduce memory requirements or reduce an amount of processing by applying machine learning operations (e.g., a convolutional neural network) to simplified (e.g., segmented) data. In some instances, stacking multiple channels of lower-dimensional images (e.g., over time) improves segmentation and/or classification by incorporating temporal information into the operations. These and other improvements to the functioning of the computer are discussed herein.”; [0039]: “FIG. 2 illustrates a pictorial flow diagram of a process 200 for capturing three-dimensional data of an object, receiving segmentation information, adapting a rendering perspective for the object, converting the three-dimensional data of the object to two-dimensional data of the object, and performing classification.”; [0041]: “At operation 214, the process can include receiving segmentation information associated with the three-dimensional data of the object. In some instances, the segmentation information can be generated according to the process 100 illustrated in FIG. 1. In some instances, the segmentation information generated in the process 100 may include segmentation information associated with the two-dimensional representation of the three-dimensional data, in which case, the two-dimensional segmentation information may be converted to three-dimensional segmentation information.”);
running a principal component analysis on a plurality of pixels of said segmentation mask (see at least [0043]: “For example, the operation 222 may include determining a center of the segmented data 220, which may include determining a “center of mass” of the data points of the segmented data 220. Next, the operation 222 may include performing a principal component analysis of the segmented data to determine eigenvectors or principal components of the segmented data 220.”);
identifying a primary axis explaining maximum variance and a secondary axis along which residuals are minimalized (see at least [0043]: “Next, the operation 222 may include performing a principal component analysis of the segmented data to determine eigenvectors or principal components of the segmented data 220. In some instances, a first principal component may correspond to an axis of “maximum stretch” or variance of the segmented data 220. In some instances, the first principal component can be selected or determined as a principal component in the (x, y) plane (e.g., a horizontal plane). Next, a second principal axis can be selected as a principal component orthogonal to the first principal component in a vertical direction (e.g., the z-direction). In some instances, an initial rendering plane may be defined by the first principal component and the second principal component, and in some instances, the rendering plane (e.g., the rendering plane 226) may be determined by rotating the initial rendering plane about the first principal component, which conceptually can be considered as increasing a height of the rendering perspective 228. In some instances, the rendering plane 226 may be determined by rotating the initial rendering plane by a predetermined angle or number of degrees, or by selecting a predetermined height or a predetermined change in height for the rendering perspective 228, for example. Further, a center of the rendering plane 226 may be associated with a center of mass of the segmented data 220, for example, following a determination of the orientation of the rendering plane 226.”);
orienting a candidate such that an axis aligns with said secondary axis (see at least (see at least [0043]: “Next, a second principal axis can be selected as a principal component orthogonal to the first principal component in a vertical direction (e.g., the z-direction). In some instances, an initial rendering plane may be defined by the first principal component and the second principal component, and in some instances, the rendering plane (e.g., the rendering plane 226) may be determined by rotating the initial rendering plane about the first principal component, which conceptually can be considered as increasing a height of the rendering perspective 228. In some instances, the rendering plane 226 may be determined by rotating the initial rendering plane by a predetermined angle or number of degrees, or by selecting a predetermined height or a predetermined change in height for the rendering perspective 228, for example. Further, a center of the rendering plane 226 may be associated with a center of mass of the segmented data 220, for example, following a determination of the orientation of the rendering plane 226.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Douillard and provide a means to obtain an image segmentation mask of said target object, running a principal component analysis on a plurality of pixels of said segmentation mask, and identifying a primary axis explaining maximum variance and a secondary axis along which residuals are minimalized, with a reasonable expectation of success, in order to reduce data dimensionality while maintaining important information and reduce memory requirements or reduce an amount of processing by applying machine learning operations [0031].
16. Claim 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) and Ku et al. (US 20220016767, hereinafter Ku) and of Prats (US 20170326728, hereinafter Prats) and in further view of Oleynik (US 20170348854, hereinafter Oleynik).
Regarding claim 11, modified Barry teaches the limitations of claim 8. Barry further teaches wherein said grasp proposal method comprises:
presenting a previously unknown object to said robot manipulator as a target object (see at least [0034]: “The image displayed on the UI 300 may include one or more objects that are present in the environment 10 (e.g., within a field of view F.sub.V for a sensor 132 of the robot 100). In some examples, the grasping system 200 or some other system of the robot 100 may be configured to classify an image in order to identify one or more objects within the image (e.g., to identify one or more graspable objects). In some implementations, the image is classified by a machine learning algorithm in order to identify the presence of one or more graspable objects in the image that correspond to one or more graspable objects within a portion of the environment 10 corresponding to the image. In particular, the sensor system 130 receives the image that corresponds to the area (e.g., environment 10) and sends the image (e.g., sensor data 134) to the grasping system 200. The grasping system 200 classifies graspable objects within the received image (e.g., sensor data 134) using a machine learning object classification algorithm. For example, the grasping system 200 may classify a piece of clothing on the ground as “laundry,” or a piece of trash on the ground as “trash.” The classification of the objects in the image may display to the user 12 on the UI 300. The UI 300 may further calibrate the received image to display for the user 12. The UI 300 allows the user 12 to select an object displayed in the two-dimensional image as a target object in order to instruct the robot 100 to perform an action upon the selected target object in the three-dimensional environment 10.”);
moving a gripper camera around said target object (see at least [0050]: “The adjuster 230 determines whether a respective grasp geometry 212 from the new set of candidate grasp geometries 212N includes a corresponding grasping score 242 that exceeds the grasping score 242 of the initial grasp geometry 212I (i.e., a score 242 that indicates a candidate grasp geometry 212N is better than the initial grasp geometry 212I). That is, the adjuster 230 receives the updated sensor data 134U and the respective grasping score 242 for the initial grasp geometry 212I and for each new candidate grasp geometry 212N.”);
acquiring, via said gripper camera, a 3-Dimensional (“3D”) representation of said target object (see at least [0025]: “The sensor system may additionally and/or alternatively generate the field of view F.sub.V with a sensor 132 mounted at or near the end-effector 150 of the arm 126 (e.g., sensor(s) 132c). The one or more sensors 132 may capture sensor data 134 that defines the three-dimensional point cloud for the area within the environment 10 about the robot 100. In some examples, the sensor data 134 is image data that corresponds to a three-dimensional volumetric point cloud generated by a three-dimensional volumetric image sensor 132. Additionally or alternatively, when the robot 100 is maneuvering about the environment 10, the sensor system 130 gathers pose data for the robot 100 that includes inertial measurement data (e.g., measured by an IMU). In some examples, the pose data includes kinematic data and/or orientation data about the robot 100, for instance, kinematic data and/or orientation data about joints J or other portions of a leg 120 or arm 126 of the robot 100. With the sensor data 134, various systems of the robot 100 may use the sensor data 134 to define a current state of the robot 100 (e.g., of the kinematics of the robot 100) and/or a current state of the environment 10 about the robot 100.”; [0036]: “For instance, the grasping system 200 generates a grasp area or grasp region corresponding to the area within the three-dimensional environment 10 where the target object is actually located in to order designate where the end-effector 150 is to grasp the target object. In particular, the grasping system 200 transforms the user-selected target object from the two-dimensional image to the grasp area on the three-dimensional point cloud of sensor data 134. By generating the grasp area, the grasping system 200 allows the selected target object from the two-dimensional image to instruct the robot 100 to grasp the target object in the three-dimensional environment 10. In some configurations, the grasping system 200 generates the grasp area by projecting a plurality of rays from the selected target object of the image onto the three-dimensional point cloud of sensor data 134, as discussed in more detail below in FIG. 2C. After determining the grasp area, the grasping system 200 determines a grasp geometry 212 for the robotic manipulator (i.e., the arm 126 of the robot 100) to grasp the target object with.”); and
associating the target object with a second object within said library of known objects (see at least [0034]: “In some implementations, the image is classified by a machine learning algorithm in order to identify the presence of one or more graspable objects in the image that correspond to one or more graspable objects within a portion of the environment 10 corresponding to the image. In particular, the sensor system 130 receives the image that corresponds to the area (e.g., environment 10) and sends the image (e.g., sensor data 134) to the grasping system 200. The grasping system 200 classifies graspable objects within the received image (e.g., sensor data 134) using a machine learning object classification algorithm. For example, the grasping system 200 may classify a piece of clothing on the ground as “laundry,” or a piece of trash on the ground as “trash.” The classification of the objects in the image may display to the user 12 on the UI 300. The UI 300 may further calibrate the received image to display for the user 12.”).
Barry fails to explicitly teach demonstrating, via an operator, a suitable grasp pose on said target object; storing said demonstration, along with a descriptor of said target object, in a memory library of known objects; and matching demonstrated grasps from said second object within said library of known objects to said target object.
However, Oleynik teaches a method and system for creating robotic humanoid movements that demonstrates, via an operator, a suitable grasp pose on said target object (see at least [0416]: “FIG. 5B is a block diagram illustrating one embodiment of the standardized chef studio 44 and robotic kitchen 50 with teach/playback process 176. The teach/playback process 176 describes the steps of capturing a chef's recipe-implementation processes/methods/skills 49 in the chef studio 44 where he/she carries out the recipe execution 180, using a set of chef-studio standardized equipment 74 and recipe-required ingredients 178 to create a dish while being logged and monitored 182.”);
storing said demonstration, along with a descriptor of said target object, in a memory library of known objects (see at least [0416]: “The raw sensor data is logged (for playback) in 182 and processed to generate information at different abstraction levels (tools/equipment used, techniques employed, times/temperatures started/ended, etc.), and then used to create a recipe-script 184 for execution by the robotic kitchen 48. The robotic kitchen 48 engages in a recipe replication process 106, whose profile depends on whether the kitchen is of a standardized or non-standardized type, which is checked by a process 186.”; [0427]: “The minimanipulation library is a command-software repository, where motion behaviors and processes are stored based on an off-line learning process, where the arm/wrist/finger motions and sequences to successfully complete a particular abstract task (grab the knife and then slice; grab the spoon and then stir; grab the pot with one hand and then use other hand to grab spatula and get under meat and flip it inside the pan; etc.”; [0440]: “The cloud computing 395 includes computer storage locations to store a task library 398a with actions, recipe, and minimanipulations; a user profile/data 398b with login information, ID, and subscriptions; a recipe meta data 398c with text, voice media, etc.; an object recognition module 398d with standard images, non-standard images, dimensions, weight, and orientations; an environment/instrumented map 398e for navigation of object positions, locations, and the operating environment; and a controlling software files 398f for storing robotic command instructions, high-level software files, and low-level software files. In another embodiment, the Internet of Things (IoT) devices can be incorporated to operate with the chef kitchen 44, the cloud computing 396 and the robotic kitchen 48.”); and
matching demonstrated grasps from said second object within said library of known objects to said target object (see at least [0447]: “These action primitives by the chef 49, as recorded by the gloves 26a, 26b, may constitute a minimanipulation 432 that take place over time slots 1, 2, 3 and 4. The recipe algorithm conversion module 404 is configured to convert the recorded recipe file from the chef studio 44 to robotic instructions for operating the robotic arms 70 and the robotic hands 72 in the robotic kitchen 28 according to a software table 434.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Oleynik and provide a means to demonstrate, via an operator, a suitable grasp pose on said target object, store said demonstration, along with a descriptor of said target object, in a memory library of known objects, and matching demonstrated grasps from said second object within said library of known objects to said target object, with a reasonable expectation of success, in order to replicate a demonstration from an expert on how to grasp various objects.
17. Claim 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Barry et al. (US 20220193894, hereinafter Barry) and Ku et al. (US 20220016767, hereinafter Ku) and in view of Hoffman et al. (US 20170217021, hereinafter Hoffman).
Regarding claim 17, modified Barry teaches the limitations of claim 15.
Barry fails to explicitly teach wherein said user interaction device comprises at least one joystick.
However, Hoffman teaches a method and apparatus for remotely operating a mobile robot with a user interaction device that comprises at least one joystick (see at least Figs. 1, 4A, and [0076]: “Referring to FIGS. 4A-4M, the OCU 100 may include a display (e.g., LCD or touch screen) 110, a keyboard, and one or more auxiliary user inputs, such as a joystick or gaming unit in communication with the computing device 102. As shown, the OCU 100 is a touch screen tablet. The OCU 100 provides a user interface of the teleoperation software application 101 that is rendered on the display 110 of the OCU 100 and allows an operator or user 10 to control the robot 200 from a distance.”).
Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Barry to incorporate the teachings of Hoffman and provide a user interaction device that comprises at least one joystick, with a reasonable expectation of success, in order to provide an alternative means of controlling the robot.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Chavez et al. (US 20200316782) teaches a method and system for robotic picking and placing unknown objects from various locations and displaying information on an input device for a human operator.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TIEN MINH LE whose telephone number is (571)272-3903. The examiner can normally be reached Monday to Friday (8:30am-5:30pm eastern time).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Khoi Tran can be reached on (571)272-6919. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/T.M.L./Examiner, Art Unit 3656
/KHOI H TRAN/Supervisory Patent Examiner, Art Unit 3656
Prosecution Insights