`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_____________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_____________________________
`
`ABB INC.,
`Petitioner,
`
`v.
`
`ROBOTICVISIONTECH, INC.,
`Patent Owner
`_____________________________
`
`Case: IPR2023-1426
`U.S. Patent No. 8,095,237
`_____________________________
`
`PETITION FOR INTER PARTES REVIEW OF U.S. PATENT NO. 8,095,237
`
`
`
`
`
`
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`
`I. 
`
`TABLE OF CONTENTS
`  MANDATORY NOTICES UNDER 37 C.F.R. § 42.8 ................................... 1 
`A.  Real Party-in-Interest ....................................................................................... 1 
`B.  Related Matters ................................................................................................ 1 
`C.  Lead and Back-up Counsel and Service Information ...................................... 1 
`II. 
`  GROUNDS FOR STANDING ........................................................................ 2 
`III.    REQUESTED RELIEF ................................................................................... 2 
`IV.    REASONS FOR THE REQUESTED RELIEF .............................................. 2 
`A.  Summary of the ’237 Patent ............................................................................ 2 
`B.  Prosecution History ......................................................................................... 4 
`V. 
`  STATUTORY GROUNDS FOR CHALLENGES ......................................... 5 
`VI.    THE CHALLENGES ARE BASED ON PRIOR ART PATENTS AND
`PRINTED PUBLICATIONS .......................................................................... 6 
`VII.   LEVEL OF ORDINARY SKILL IN THE ART ............................................. 7 
`VIII.  CLAIM CONSTRUCTION ............................................................................ 8 
`IX.    STATE OF THE PRIOR ART ........................................................................ 9 
`A.  Corke ................................................................................................................ 9 
`X. 
`  GROUND 1: CLAIMS 1-4, 6-10, 17-20, AND 24-28 ARE
`UNPATENTABLE AS OBVIOUS OVER CORKE IN VIEW OF THE
`KNOWLEDGE OF A POSITA ..................................................................... 11 
`A.  Claim 1 ........................................................................................................... 11 
`B.  Claim 2 ........................................................................................................... 22 
`C.  Claims 3 and 4 ............................................................................................... 26 
`D.  Claim 6 ........................................................................................................... 30 
`E.  Claim 7 ........................................................................................................... 31 
`F.  Claim 8 ........................................................................................................... 31 
`G.  Claim 9 ........................................................................................................... 34 
`H.  Claim 10 ......................................................................................................... 36 
`I.  Claims 17, 24, and 28 .................................................................................... 37 
`J.  Claim 18 ......................................................................................................... 39 
`
`i
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`K.  Claim 19 ......................................................................................................... 41 
`L.  Claims 20 and 25 ........................................................................................... 43 
`M.  Claims 26 and 27 ........................................................................................... 55 
`XI.    GROUND 2: CLAIMS 5, 12-16, AND 21-24 ARE UNPATENTABLE AS
`OBVIOUS OVER CORKE IN VIEW OF WEI-I ......................................... 56 
`A.  Wei-I .............................................................................................................. 56 
`B.  Motivation to Combine Corke and Wei-I ...................................................... 57 
`C.  Claims 5 and 12 ............................................................................................. 59 
`D.  Claim 13 ......................................................................................................... 64 
`E.  Claim 14 ......................................................................................................... 65 
`F.  Claim 15 ......................................................................................................... 67 
`G.  Claim 16 ......................................................................................................... 68 
`H.  Claim 21 ......................................................................................................... 73 
`I.  Claims 22 and 23 ........................................................................................... 76 
`XII.   DISCRETIONARY DENIAL UNDER § 314 OR § 325(D) IS NOT
`WARRANTED .............................................................................................. 79 
`A.  § 314 .............................................................................................................. 79 
`B.  § 325(d) .......................................................................................................... 81 
`
`
`
`
`
`
`ii
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Cases
`Apple Inc. v. Fintiv, Inc.,
`IPR2020-00019, Paper 11 (PTAB March 20, 2020) .......................................... 79
`Apple Inc. v. Fintiv, Inc.,
`IPR2020-00019, Paper 15 (PTAB May 13, 2020) ............................................. 79
`Becton, Dickinson & Co. v. B. Braun Melsungen AG,
`IPR2017-01586, Paper 8 (P.T.A.B. Dec. 15, 2017) ........................................... 81
`Bristol-Myers Squibb Co. v. Ben Venue Lab’ys, Inc.,
`246 F.3d 1368 (Fed. Cir. 2001) .......................................................................... 11
`Harris Corp. v. Ericsson Inc.,
`417 F.3d 1241 (Fed. Cir. 2005) ............................................................................ 8
`Koninklijke Philips N.V. v. Google LLC,
`948 F.3d 1330 (Fed. Cir. 2020) .................................................................... 14, 15
`KSR Int’l Co. v. Teleflex, Inc.,
`550 U.S. 398 (2007) ................................................................................ 14, 60, 61
`
`Sand Revolution II, LLC v. Continental Intermodal Group - Trucking
`LLC,
`IPR2019-01393, Paper 24 (PTAB June 16, 2020) ............................................. 80
`TecSec, Inc. v. Int’l Bus. Machs. Corp.,
`731 F.3d 1336 (Fed. Cir. 2013) ............................................................................ 8
`Statutes
`35 U.S.C. §102 ........................................................................................................... 6
`35 U.S.C. § 103 ................................................................................................ 5, 7, 14
`35 U.S.C. §112(6) ...................................................................................................... 9
`35 U.S.C. § 314 ........................................................................................................ 79
`35 U.S.C. §315(e)(2) ................................................................................................ 80
`
`iii
`
`

`

`IPR2023-01426
`IPR2023-01426
`U.S. Patent No. 8,095,237
`U.S. Patent No. 8,095,237
`35 U.S.C. §316(a)(11) .............................................................................................. 79
`35 U.S.C. §316(a)(11)sccccssessccsssescsssesscsssssecssssccessusscssssesessseesessuesessssssesssessersneeesssesecs 79
`§ 325(d) .............................................................................................................. 79, 81
`§ 325(d) cecccccccsccescssssesccenssssececcerssscesceessssesecenssssseceesssssesecenssuseseesansusessecensneensesenee 79, 81
`
`
`
`
`
`iv
`iv
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`PETITIONER’S EXHIBIT LIST
`
`Exhibit Description
`
`1001
`
`U.S. Patent No. 8,095,237 (the “’237 Patent”)
`
`1002
`
`File History of U.S. Patent No. 8,095,237
`
`1003
`
`1004
`
`1005
`
`1006
`
`Declaration of Seth Hutchinson, Ph.D. Regarding Invalidity of U.S.
`Patent No. 8,095,237
`
`Peter I. Corke, Visual Control of Robots: High-Performance Visual
`Servoing (Aug. 1996) (“Corke”)
`
`Guo-Qing Wei, et al., “Active Self-calibration of Robotic Eyes and
`Hand-eye Relationships with Model Identification,” TRANSACTIONS
`ON ROBOTICS AND ANIMATION, vol. 14, No. 1, Feb 1998 (“Wei-I”)
`
`“Multisensory Visual Servoing by a Neural Network,” Guo-Qing
`Wei et al., IEEE Transactions on Systems, Man and Cybernetics,
`April 1999 (“Wei-II”).
`
`1007
`
`RoboticVISIONTech, Inc. v. ABB Inc., No. 1:22-cv-01257,
`Summons in a Civil Action (served Sept. 23, 2023).
`
`1008
`
`U.S. Patent No. 4,146,924 to Birk et al.
`
`1009
`
`U.S. Patent No. 5,959,425 to Bieman et al.
`
`1010
`
`Declaration of Sylvia Hall-Ellis, Ph.D.
`
`1011
`
`Docket Sheet, RoboticVISIONTech, Inc. v. ABB Inc., Case No.
`1:22-cv-01257-GBW (D. Del.)
`
`v
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`
`1012
`
`Curriculum Vitae of Seth Hutchinson, Ph.D.
`
`1013
`
`Scheduling Order, RoboticVISIONTech, Inc. v. ABB Inc., Case No.
`1:22-cv-01257-GBW (D. Del.)
`
`
`
`vi
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`I. MANDATORY NOTICES UNDER 37 C.F.R. § 42.8
`A. Real Party-in-Interest
`
`ABB Inc. (“ABB”) is the Petitioner and real party-in-interest.
`
`B. Related Matters
`
`U.S. Patent No. 8,095,237
`
`(the “’237 Patent”)
`
`is asserted
`
`in
`
`RoboticVISIONTech, Inc. v. ABB Inc., Case No. 1:22-cv-01257-GBW (D. Del.)
`
`(“Related Litigation”).
`
`C. Lead and Back-up Counsel and Service Information
`
`Lead Counsel
`Andrew R. Sommer
`USPTO Reg. No. 53,932
`Greenberg Traurig, LLP
`1750 Tysons Boulevard
`Suite 1000
`McLean, VA 22102
`Phone: 703-749-1370
`Fax: 703-749-1301
`sommera@gtlaw.com
`Backup Counsel
`Trenton A. Ward
`USPTO Reg. No. 59,157
`Greenberg Traurig, LLP
`333 Piedmont Road NE
`Suite 250
`Atlanta, GA 3035
`Phone: 678-553-2100
`Fax: 678-55302212
`Trenton.ward@gtlaw.com
`
`Backup Counsel
`Benjamin Schladweiler (pro hac vice
`forthcoming)
`Greenberg Traurig, LLP
`222 Delaware Avenue
`Suite 1600
`Wilmington, DE 19801
`Phone: 302-661-7352
`Fax: 302-661-7360
`schladweilerb@gtlaw.com
`Backup Counsel
`Kathryn E. Albanese
`USPTO Reg. No. 78,153
`Greenberg Traurig, LLP
`One Vanderbilt Avenue
`New York, NY 10017
`Phone: 212-801-6533
`Fax: 212-801-6200
`Katie.albanese@gtlaw.com
`
`
`Please address all correspondence to lead and back-up counsel at
`
`ABB-RVT-LITIGATION@gtlaw.com. Petitioner consents to electronic service.
`
`1
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`II. GROUNDS FOR STANDING
`Petitioner certifies that the ’237 Patent is available for inter partes review and
`
`that Petitioner is not barred or estopped from requesting inter partes review
`
`challenging the claims on the grounds identified in this Petition.
`
`III. REQUESTED RELIEF
`Petitioner requests institution of inter partes review of Claims 1-10 and 12-28
`
`(“Challenged Claims”) of the ’237 Patent, and cancelation of those claims as
`
`unpatentable.
`
`IV. REASONS FOR THE REQUESTED RELIEF
`For the reasons discussed in this Petition, the Board should institute trial and
`
`find the Challenged Claims unpatentable.
`
`A. Summary of the ’237 Patent
`
`The ’237 Patent describes the use of machine vision for 3D pose estimation.
`
`The methods of the ’237 Patent include three steps: “a) calibration of the camera; b)
`
`teaching the features on the object; and c) finding the three-dimensional pose of the
`
`object.” EX1001, 2:60-67.
`
`Figure 1 of the ’237 Patent depicts a “vision-guided robot” 10 with a
`
`manipulating arm 12 on which camera 16 and tool 14—designed to manipulate a
`
`target object—are mounted. EX1001, 2:29, 2:53-59.
`
`2
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`
`
`
`The first step is to calibrate the camera. The ’237 Patent describes three
`
`well-known types of calibration: (1) intrinsic calibration, which involves finding the
`
`“camera intrinsic parameters” describing “how the camera forms an image,”
`
`including the focal length of the camera, a radial distortion coefficient, coordinates
`
`of the center of radial lens distortion, and a scale factor, (2) extrinsic calibration,
`
`which involves finding the camera’s position and orientation (i.e., “pose”) in the
`
`world coordinate frame and (3) hand-eye calibration, which involves finding the
`
`position and orientation of the camera “relative to the tool of the robot.” EX1001,
`
`3:36-38, 3:56-67, 4:1-10, 5:51-65, 7:19-25, 7:48-49, 8:11-14, 8:30-39, 9:25-32,
`
`9:44-65; EX1003, ¶¶24-41.
`
`3
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`B. Prosecution History
`
`The ’237 Patent was filed as Application No. 10/634,874 (“the ’874
`
`Application”) on August 6, 2003 as a continuation-in-part of Application No.
`
`10/153,680, which claims the benefit of a Canadian patent application filed on
`
`January 31, 2002. EX1001. The earliest possible effective filing date of the ’237
`
`Patent claims is January 31, 2002.
`
`In eight years of prosecution, the Office rejected the claims in six Office
`
`Actions. During prosecution, Patentee canceled the original claims and presented
`
`new claims 33-61. EX1002, pp.64-65. These claims were rejected as anticipated by
`
`Wei, et al., “Multisensory Visual Servoing by a Neural Network, IEEE (1999)
`
`(“Wei-II”).1 Patentee sought to distinguish Wei-II: “Importantly…Wei[-II]…is
`
`directed to the use of a stereo pair of cameras” but the claimed invention “employ[s]
`
`single camera three-dimensional (3-D) vision for robotic guidance.” EX1002, p.129.
`
`“It is the problems associated with stereo vision based systems and other
`
`multi-camera systems which are addressed by the various single camera
`
`embodiments described in Applicants’ specification and claims.” EX1002, p.129.
`
`Patentee argued that Wei-II’s techniques “are not directly applicable” to the claimed
`
`
`1 Wei-I and Wei-II were coauthored by Wei fourteen months apart, but disclose
`completely different methods. Whereas Wei-I discloses a method for “Active
`Calibration of Robotic Eyes,” Wei-II discloses a visual servoing method that “avoids
`all such calibrations.” EX1005, p.1; EX1006, p.276.
`
`4
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`“single camera vision based methods and apparatus.” EX1002, p.130. Wei-II also
`
`sought to “avoid calibration,” and the ’237 Patent’s independent claims recited
`
`calibration steps. EX1002, p.130. On this basis, the Examiner allowed the claims.
`
`EX1002, p.193.
`
`Several rejections followed, including obviousness rejections based on U.S.
`
`Patent No. 4,942,539 to McGee in view of Wei-II, and McGee in view of U.S. Patent
`
`Application Publication No. 2002/0159628 to Matusik. EX1002, pp.858-860. In
`
`response to the last of those rejections, Patentee argued that for application claim 50,
`
`the prior art failed to show “determining an object space-to-camera space
`
`transformation for the target object based at least in part on a position of at least
`
`some of the located features using only the single captured image.” EX1002, p.945.
`
`V.
`
`STATUTORY GROUNDS FOR CHALLENGES
`Petitioner presents the following challenges:
`
`No.
`
`Claims
`
`Ground
`
`1
`
`2
`
`1-4, 6-10,
`17-20, 24-28
`
`5, 12-16, 21-
`24
`
`Obviousness over Peter I. Corke, Visual
`Control of Robots: High-Performance
`Visual Servoing (Aug. 1996) (“Corke”) in
`view of the knowledge of a POSITA
`
`Obviousness over Corke in view of Guo-
`Qing Wei, et al., “Active Self-calibration of
`Robotic Eyes and Hand-eye Relationships
`with Model Identification,” TRANSACTIONS
`ON ROBOTICS AND ANIMATION, vol. 14, No.
`1, Feb 1998 (“Wei-I”) and the knowledge of
`a POSITA
`
`5
`
`Statute
`
`§103(a)
`
`§103(a)
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`VI. THE CHALLENGES ARE BASED ON PRIOR ART PATENTS AND
`PRINTED PUBLICATIONS
`Based on the references provided herein, Petitioner assumes, without
`
`admitting, that the claims of the ’237 Patent are entitled to a filing date of January
`
`31, 2002. Corke2 was published in 1996 and was available to those working in the
`
`field no later than July 9, 1997 because, at least by this date, Corke was cataloged
`
`by subject matter in various databases including the OCLC bibliographic database
`
`and accessible to the public in the British Library and the Morgan Library at
`
`Colorado State University. EX1010, ¶¶40-45. Corke was distributed and accessible
`
`to the public based on (1) the print copy in the Morgan Library and Colorado State
`
`at least of July 9, 1997, (2) that it was cataloged by subject matter on or around that
`
`date, (3) and was widely held by libraries and was cataloged by subject matter on
`
`the OCLC bibliographic database. EX1010, ¶¶40-45. Regardless of which of these
`
`dates is the date of first publication, Corke is a prior art printed publication under
`
`§102(b). Corke is analogous to the ’237 Patent because it is in the same field as the
`
`’237 Patent—vision guided robotics. EX1001, 1:13-15; EX1004, p.3 (discussing
`
`background of robotic vision).
`
`
`2 The version of Corke submitted with the Petition was obtained from the Morgan
`Library at Colorado State University. EX1010, ¶39.
`
`6
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`Wei-I3 was included in the February 1998 edition of Transactions on Robotics
`
`and Animation. EX1005; EX1010, ¶46. Wei-I is expressly referenced in the ’237
`
`Patent specification and is Applicant Admitted Prior Art. EX1001, 8:30-39, 6:6-7:2.
`
`Specifically, the ’237 Patent states that its “algorithm is developed using the
`
`approach described in” Wei-I. EX1001, 6:6-7:2. Wei-I was distributed and
`
`accessible to the public based on (1) the date stamp on the cover page showing that
`
`it was received in the Linda Hall Library on or about February 18, 1998, (2) that it
`
`was cataloged by subject matter on or around that date, (3) and was widely held by
`
`libraries and was cataloged by subject matter on the OCLC bibliographic database.
`
`EX1010, ¶¶47-52. Wei-I was thus publicly accessible on or shortly after February
`
`18, 1998, making it prior art under at least §102(b). Wei-I is analogous to the ’237
`
`Patent because it is in the same field as the ’237 Patent—vision guided robotics.
`
`EX1001, 1:13-15; EX1005, p.1 (referring to “camera self-calibration” achieved in
`
`“robotics and computer vision.”).
`
`VII. LEVEL OF ORDINARY SKILL IN THE ART
`As of January 31, 2002, a POSITA would have had a Bachelor’s degree in
`
`robotics, mechanical engineering, computer science, electrical engineering, or an
`
`equivalent, and at least three years of professional experience working in the field
`
`
`3 The version of Wei-I submitted with the Petition was obtained from the Linda Hall
`Library of Science, Engineering & Technology. EX1010, ¶46.
`
`7
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`of computer vision or three years of graduate education, including a focus on
`
`computer vision applications. The POSITA would have knowledge about machine
`
`vision and camera calibration techniques. EX1003, ¶¶77-78.
`
`VIII. CLAIM CONSTRUCTION
`For this Petition only, Petitioner relies upon the definitions provided in the
`
`’237 Patent, including for “object space” as “a reference frame defined with respect
`
`to, and therefore rigid to, the object” and “training space” as “a reference frame
`
`defined with respect to a point on the calibration template, and aligned to its main
`
`axis.” EX1001, 3:3-34.
`
`Claims 20-28 recite “means for calibrating the camera” and “means for
`
`estimating a pose of a target object.” EX1001, 13:41-14:65. The presumption that
`
`these are means-plus-function claims can be overcome if the claims recite structural,
`
`rather than functional, language. TecSec, Inc. v. Int’l Bus. Machs. Corp., 731 F.3d
`
`1336, 1347 (Fed. Cir. 2013). “[I]t is sufficient if the claim term is used in common
`
`parlance or by persons of skill in the pertinent art to designate structure.” Id. The
`
`Federal Circuit has established
`
`that
`
`the corresponding structure
`
`for a
`
`“computer-implemented means-plus-function” is “the algorithm.” Harris Corp. v.
`
`Ericsson Inc., 417 F.3d 1241, 1253 (Fed. Cir. 2005).
`
`In the ’237 Patent, the functional language provided in Claims 20 and 28 for
`
`the “means for calibrating the camera” is the calibration algorithm itself, as the claim
`
`8
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`itself recites “determining a set of intrinsic parameters” and “determining a set of
`
`extrinsic parameters” “from at least one of the images of the calibration object,”
`
`which results in performing the calibration algorithm recited in the specification.
`
`EX1001, Fig. 5; EX1003, ¶88. The same is true with the recited “means for
`
`estimating the pose of a target object” because completing the claimed steps of
`
`“capturing a two-dimensional image of a volume containing a target object,”
`
`“locating” a number of “features in the captured image,” and “determining an object
`
`space-to-camera space transformation,” results in performing the pose estimation
`
`algorithm recited in the specification. EX1003, ¶89; EX1001, Fig. 7. Therefore,
`
`Claims 20-28 are not means-plus-function claims and no construction is necessary.
`
`A POSITA reading the specification of the ’237 Patent would have understood that
`
`Claims 20-28 provide sufficient structure to perform the algorithm cited therein;
`
`thus, these terms do not invoke 35 U.S.C. §112(6). EX1003, ¶90.
`
`IX. STATE OF THE PRIOR ART
`A. Corke
`
`Corke is a textbook published in 1996 about the “application of high-speed
`
`machine vision for closed-loop position control, or visual servoing, of a robot
`
`manipulator.” EX1004, p.ix; EX1003, ¶¶79-81. Corke explains that “visual servoing
`
`involves the use of one or more cameras and a computer vision system to control the
`
`position of the robot’s end-effector relative to the workpiece as required by the task.”
`
`9
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`EX1004, p.1. In Figure 5.4, Corke discloses a robot having a single camera to
`
`implement a method for 3D pose estimation:
`
`
`
`EX1004, Figure 5.4.
`
`Corke explains that a key component of visual servoing systems is the use of
`
`vision sensing and a closed “feedback” loop through which the system (for each
`
`video frame) takes an image of the target object, extracts features from the object,
`
`determines the “relative pose between the camera and the target” using that single
`
`image and “knowledge of the geometric relationship between several feature points
`
`on the target,” and sends the pose estimation to the robot, causing the robot to adjust
`
`the position of the robot’s end-effector relative to the object. EX1004, pp.3, 152-55.
`
`
`
`
`
`10
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`X. GROUND 1: CLAIMS 1-4, 6-10, 17-20, AND 24-28 ARE
`UNPATENTABLE AS OBVIOUS OVER CORKE IN VIEW OF THE
`KNOWLEDGE OF A POSITA
`A. Claim 1
`[1.0] A useful in three-dimensional pose estimation for use with a single camera
`mounted to a movable portion of a robot, the method comprising:
`To the extent the Board construes the preamble as limiting, Corke discloses
`
`the preamble. EX1003, ¶¶93-96. A preamble is limiting if it gives “life, meaning,
`
`and vitality” to a claim. Bristol-Myers Squibb Co. v. Ben Venue Lab’ys, Inc., 246
`
`F.3d 1368, 1373 (Fed. Cir. 2001). Here, the recitation in the preamble of “a single
`
`camera mounted to a moveable portion of a robot” is limiting in view of Patentee’s
`
`reliance upon this limitation for allowance. As detailed above, Patentee
`
`distinguished the claims over the prior art as “directed to methods and apparatus that
`
`employ single camera three-dimensional (3-D) vision for robotic guidance.”
`
`EX1002, p.129.
`
`As to the method recited in the preamble of Claim 1, Corke discloses methods
`
`for 3D pose estimation for which “[t]he camera may be fixed, or mounted on the
`
`robot’s end-effector in which case there exists a constant relationship, t6xc, between
`
`the pose of the camera and the pose of the end-effector.” EX1004, pp.152, 207
`
`(disclosing a “single camera” mounted on robot wrist). Corke’s techniques designed
`
`to estimate the “3D position and orientation”—i.e., “pose”—of an object are useful
`
`11
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`in three-dimensional pose estimation.4 EX1004, p.3 n.2; EX1003, ¶95. Placing the
`
`camera on the robot’s arm—corresponding to the claimed moveable portion of a
`
`robot—provides certain advantages, including “the ability to avoid occlusion,
`
`resolve ambiguity and increase accuracy, by directing its attention.” EX1004, pp.3,
`
`166. Figures 4.11, 5.1, 6.10 and 6.11 of Corke illustrate “a single camera mounted
`
`to a moveable part of the robot”:
`
`
`
`
`
`
`
`
`4 Underlined text denotes language from the claims, unless otherwise noted.
`
`12
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`
`
`
`
`
`EX1004, pp.148, 152, 186-87; EX1003, ¶95. Corke in view of the knowledge of a
`
`POSITA discloses or renders obvious Preamble [1.0]. EX1003, ¶¶93-96.
`
`[1.1] capturing a two-dimensional image of a volume containing a target object;
`Corke teaches a method that captur[es] a two-dimensional image of a volume
`
`containing a target object. EX1003, ¶¶97-101. Corke discloses that “[t]he camera
`
`contains a lens which forms a 2D projection of the scene on the image plane where
`
`the sensor is located.” EX1004, pp. 152, 159 (“A broad definition of position-based
`
`servoing…includes methods based on analysis of 2D features”), 159 (disclosing
`
`method to “determine the 3D relative pose of an object…from 2D image plane
`
`coordinates”); EX1003, ¶98.
`
`As shown in Figure 5.4, Corke discloses a system that uses a video camera to
`
`capture and process multiple two-dimensional images of a target object, one image
`
`at a time at the “camera frame rate,” which is “effectively the sample rate in a visual
`
`13
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`servoing system.” EX1004, p.103, 73 (“the camera’s shutter acts as the sampler in a
`
`visual control loop”):5
`
`
`
`EX1004, p.155; EX1003, ¶99.
`
`The claim requires that the target object reside in a volume, which Corke
`
`teaches is the space from which the image of the target object is captured. EX1004,
`
`pp.152-153 (referring to “an image of the target” as “the object of interest, that is,
`
`the object that will be tracked,” so that “features” can be “extracted from the
`
`image”); EX1003, ¶100. In KSR Int’l Co. v. Teleflex, Inc., the Supreme Court held
`
`the obviousness analysis should consider the “background knowledge possessed by
`
`a person having ordinary skill in the art.” 550 U.S. 398 (2007).6 A POSITA would
`
`
`5 The portions of Corke cited herein are from a single embodiment referenced in
`both Chapter 3 (“Fundamentals of image capture”) and Chapter 4 (“Machine vision”
`concepts), and used as a baseline for the position-based visual servoing methods
`discussed in Chapter 5. EX1004.
`6 In Koninklijke Philips N.V. v. Google LLC, the Federal Circuit determined that
`“under 35 U.S.C. § 103, the obviousness inquiry turns not only on the prior art, but
`
`
`14
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`have understood that all 3D objects occupy a volume. EX1003, ¶100. Corke
`
`discloses that “[t]he camera contains a lens which forms a 2D projection of the scene
`
`on the image plane,” and “each point on the image plane corresponds to a ray in 3D
`
`space.” EX1004, pp.152. Accordingly, Corke’s
`
`teaching of capturing a
`
`two-dimensional image of a target object by a single camera would have included
`
`capturing a two-dimensional image of a volume containing a target object. EX1003,
`
`¶100. Corke in view of the knowledge of a POSITA renders obvious Element [1.1].
`
`EX1003, ¶¶97-101.
`
`[1.2] locating a number of features in the captured image of the target object;
`Corke discloses locating a number of features in the captured image of the
`
`target object. EX1004, pp.136, 168; EX1003, ¶¶102-107. For example, Corke
`
`discloses a method of “locating features” in each captured image where
`
`“[d]etermining the initial location of features requires the entire image to be
`
`searched.” EX1004, p.136, 168; EX1003, ¶103. Corke discloses that “[a] good
`
`feature is one that can be located unambiguously” such as “a hole in a gasket [] or
`
`a contrived pattern.” EX1004, pp.153; EX1003, ¶103.
`
`
`whether the differences between the claimed invention and the prior art are such that
`the claimed invention as a whole would have been obvious . . . to a person having
`ordinary skill in the art to which the claimed invention pertains.” 948 F.3d 1330,
`1337 (Fed. Cir. 2020) (affirming PTAB’s determination of unpatentability based
`upon cited prior reference in view of knowledge of POSITA).
`
`15
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`Corke also discloses that in “position based control, features are extracted
`
`from the image” of a target object to “determine the pose of the target with respect
`
`to the camera.” EX1004, p.152-53 (“The task in visual servoing is to control the pose
`
`of the robot’s end-effector, xT6, using visual information, features, extracted from
`
`the image ”). EX1003, ¶104. Figure 5.4 shows the “Image feature extraction” step
`
`disclosed in Corke:
`
`
`
`EX1004, p.155. A POSITA would have understood that Corke teaches that features
`
`are located and processed one image at a time in view of Corke’s express disclosure
`
`that “features are extracted from the image.” EX1004, p.153, 152 (“features”
`
`“extracted from the image”); Figure 5.4 (showing “Image feature extraction”).
`
`EX1003, ¶105. As an example of single-image processing, Corke teaches use of the
`
`Datacube DIGIMAX for acquiring an individual image from the camera and
`
`processing this individual image to locate features, stating “[t]he incoming analog
`
`16
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`video signal is digitized to form a 512x512 pixel digital image,” after which various
`
`features are detected, such as “moments up to second order, perimeter and bounding
`
`box.” EX1004, p.180; EX1003, ¶105.
`
`Corke also discloses that when machine vision is used for “part inspection and
`
`quality control,” the process involves “extraction” of “features from the image[.]”
`
`EX1004, p.123. By disclosing that “features are extracted,” Corke teaches the
`
`POSITA that a number of features are located in the captured image of the target
`
`object. EX1003, ¶106.
`
`Corke in view of the knowledge of a POSITA renders obvious Element [1.2].
`
`EX1003, ¶¶102-107.
`
`[1.3] determining by a processor an object space-to-camera space transformation
`for the target object based at least in part on a position of at least some of the
`located features using only the single captured image and an algorithm that
`employs a known or determinable physical relationship between at least some of
`the located features.
`Corke discloses determining by a processor an object space-to-camera space
`
`transformation for the target object utilizing the position of the features from a single
`
`image. EX1003, ¶¶108-116. The ’237 Patent provides that “object space” is a 3D
`
`coordinate frame “defined with respect to a point on, and therefore rigid to, the
`
`object,” “camera space” is a coordinate frame rigid to the camera, and
`
`“transformation” is the “three-dimensional rotation & translation between two
`
`spaces” or coordinate frames. EX1001, 3:1-34.
`
`17
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`Corke discloses that in “position based control, features are extracted from
`
`the image” of the target object “and used in conjunction with a geometric model of
`
`the target to determine the pose of the target with respect to the camera.” EX1004
`
`p.153. A POSITA would have understood that determining the target’s pose with
`
`respect to the camera is equivalent to the claimed step of determining the object
`
`space-to-camera space transformation. EX1003, ¶110. This object space-to-camera
`
`space transformation is depicted in Figure 5.1 of Corke, where P designates the
`
`target object and cxt denotes “the relative pose between the camera and the target,”
`
`i.e., the transformation between the camera and the target object—the “object of
`
`interest”:
`
`EX1004 pp.152, 138; EX1003, ¶110.
`
`18
`
`
`
`

`

`IPR2023-01426
`U.S. Patent No. 8,095,237
`A POSITA would have understood that Corke’s references t