`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`
`
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Shenzhen Zhiyi Technology Co. Ltd., d/b/a iLife,
`Petitioner,
`
`v.
`
`iRobot Corp.,
`Patent Owner
`
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2017-02078
`Patent 7,155,308
`
`
`
`
`
`
`
`
`
`
`
`
`PATENT OWNER’S
`PRELIMINARY RESPONSE
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`
`
`TABLE OF CONTENTS
`INTRODUCTION ......................................................................................... 1
`I.
`BACKGROUND ............................................................................................ 2
`II.
`III. LEVEL OF ORDINARY SKILL IN THE ART ........................................ 3
`IV. CLAIM CONSTRUCTION .......................................................................... 5
`A.
`“re-direct[ing]” (claims 1 and 19) ......................................................... 6
`1.
`The Intrinsic Record Confirms that “re-direct[ing]” Should be
`
`Construed as “changing a direction of travel” ............................ 7
`
`2.
`
`Extrinsic Evidence Supports Patent Owner’s Construction ..... 10
`
`V. GROUNDS 1-4 SHOULD BE DENIED IN FULL BECAUSE THE
`CITED REFERENCES FAIL TO TEACH OR SUGGEST “RE-
`DIRECT[ING] THE AUTONOMOUS ROBOT” .................................... 11
`VI. THE PETITION’S ANALYSIS OF CLAIM 19 IS INCOMPLETE ...... 16
`A.
`THE PETITION’S ANALYSIS OF ELEMENTS 19(A) – (B) .......... 16
`B.
`THE PETITION’S ANALYSIS OF CLAIM 19(C) ........................... 18
`VII. DRUNK-106’S SENSORS ARE NOT PROXIMATE TO A WHEEL
`(CLAIM 28) .................................................................................................. 21
`VIII. GROUNDS 2 AND 4 AGAINST CLAIMS 3 AND 17 BASED ON LEE-
`291 SHOULD BE DENIED ........................................................................ 27
`A.
`Petitioner Improperly Shifts Its Obviousness Theory For Claim 3 .... 27
`B.
`Petitioner’s Proposed Obviousness Theory For Claim 17 Is Rooted In
`Hindsight-Driven Speculation ............................................................. 31
`IX. THE PETITION IS REPLETE WITH SUBSTANTIVE
`DEFICIENCIES .......................................................................................... 35
`
`
`
`i
`
`
`
`A.
`
`B.
`C.
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Petitioner Offers Scant Evidence That The Jones-MR And Everett
`References Were Publicly Available Before The ’308 Patent ............ 36
`Petitioner’s Attorney Arguments Are Unsupported By Evidence ...... 38
`Grounds 3 And 4 Based On Everett Are Not Fully Set Forth In The
`Petition ................................................................................................. 38
`THE PETITION PRESENTS HORIZONTALLY REDUNDANT
`GROUNDS BASED ON JONES-MR AND EVERETT .......................... 40
`XI. CONCLUSION ............................................................................................ 42
`
`
`X.
`
`
`
`ii
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`LIST OF EXHIBITS
`
`IR2003
`
`Description
`Exhibit No.
`American Heritage Dictionary, 4th Ed. (2000) (excerpted portions)
`IR2001
`IR2002 Merriam-Webster’s Collegiate Dictionary, 10th Ed. (1997)
`(excerpted portions)
`Certain Robotic Vacuum Cleaning Products and Components
`Thereof Such as Spare Parts, Investigation No. 337-TA-1057,
`Claim Construction Order No. 27 (Nov. 9, 2017)
`Yongjie, et al., A Simulation Platform For The Cooperation Of
`Distributed Robot Team, Proceedings of the IEEE International
`Conference on Mechatronics & Automation (July 2005)
`U.S. Pat. Pub. No. 2008/0039974 (“Sandin”)
`U.S. Pat. No. 6,294,084 (“Henkin”)
`
`IR2004
`
`IR2005
`IR2006
`
`
`
`iii
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`TABLE OF AUTHORITIES
`
`CASES
`
`PAGES
`
`Apple Inc., v. Contentguard Holdings, Inc., IPR2015-00449.................................. 34
`
`C.R. Bard, Inc. v. Medline Industries, Inc., IPR2015-00511................................... 34
`
`Cisco Systems, Inc. v. Constellation Techs. LLC, IPR2014-01179 ......................... 34
`
`In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006) ...................................................... 20
`
`In re Man Mach. Interface Techs. LLC, 822 F.3d 1282, 1287 (Fed. Cir. 2016) ....... 5
`
`In re Nuvasive, 842 F.3d 1376, 1382 (Fed. Cir. 2016) ............................................ 20
`
`Ineos USA LLC v. Berry Plastics Corp., 783 F.3d 865, 870-71 (Fed. Cir. 2015) ... 34
`
`KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007) .................................. 20, 34
`
`Liberty Mutual Ins. Co. v. Progressive Casualty Ins. Co., CBM2012-00003 ......... 40
`
`Microsoft Corp. v. Corel Software, LLC, IPR2016-01083 ...................................... 37
`
`Oil States Energy Servs. LLC v. Greene’s Energy Group, LLC, Case No. 16-712,
`certiorari granted (U.S. Jun. 12, 2017) ................................................................ 1
`One World Techs, Inc. v. Chamberlain Grp., Inc., IPR2016-01772 ....................... 20
`
`ServiceNow, Inc. v. Hewlett-Packard Co., IPR2015-00707 .................................... 37
`
`Verlander v. Garner, 348 F.3d 1335, 1371 (Fed. Cir. 2003) ................................... 10
`
`VirnetX Inc. v. Apple Inc., 665 F. App’x 880, 887-88 (Fed. Cir. 2016) ............ 34, 35
`
`Zimmer Biomet Holdings, Inc., v. Four Mile Bay, LLC, IPR2016-00011 ............... 10
`
`
`
`
`
`iv
`
`
`
`
`STATUTES
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`
`PAGES
`
`35 U.S.C. § 323 .......................................................................................................... 1
`
`35 U.S.C. § 102(b) ................................................................................................... 36
`
`REGULATIONS
`
`PAGES
`
`37 C.F.R. § 42.100(b) ................................................................................................ 5
`
`37 C.F.R. § 42.104(b)(4)-(5) .................................................................................... 39
`
`
`
`
`
`v
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`
`I.
`
`INTRODUCTION
`The Board should deny institution because (1) the Petition fails to
`
`demonstrate that the prior art teaches “re-directing” an autonomous robot as
`
`required by independent claims 1 and 19, (2) the Petition provides an incomplete
`
`analysis of claim 19, (3) the Petition fails to demonstrate that the prior art renders
`
`obvious dependent claims 3, 17, and 28, and (4) the Petition is substantively
`
`deficient in its analysis of individual claim elements and its scant evidence of
`
`public availability of purported non-patent prior art. Each of these reasons is fatal
`
`and warrants denial of institution.1
`
`
`
`1 Arguments presented herein are presented without prejudice to presenting
`
`additional arguments in a later response should the Board institute inter partes
`
`review. No waiver is intended by Patent Owner and no waiver attaches to
`
`arguments not presented in a patent owner’s preliminary response. 35 U.S.C.
`
`§ 323. Moreover, as the Board is aware, the United States Supreme Court is
`
`currently considering certain aspects of the inter partes review process in Oil
`
`States Energy Servs. LLC v. Greene’s Energy Group, LLC, Case No. 16-712,
`
`certiorari granted (U.S. Jun. 12, 2017). Patent Owner respectfully reserves its right
`
`to timely request permission to brief the implications of that decision on this
`
`proceeding.
`
`
`
`1
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Additionally, without justification, the Petition presents horizontally
`
`redundant grounds between the Jones-MR (Grounds 1 and 2) and Everett grounds
`
`(Ground 3 and 4). For this additional reason, the Board should deny at least one of
`
`the redundant sets of grounds.
`
`II. BACKGROUND
`The ’308 patent is directed to “an obstacle detection system for an
`
`autonomous robot, such as an autonomous cleaning robot.” Ex. 1001, 1:15-17.
`
`Specifically, the ’308 patent provides a “detection system which prevents an
`
`autonomous cleaning robot from driving off a stair or obstacle which is too high.”
`
`Id., 54-57.
`
`To improve stair or obstacle detection, the inventors of the ’308 patent
`
`purposed specially configured sensors that enabled the robot to detect stairs while
`
`traveling on a variety of surfaces, and to re-direct upon encountering stairs or other
`
`drops that the robot could fall over. Id., 5:31-43, 5:63-6:5. For example, the ’308
`
`patent describes sensor subsystems that include an optical emitter and a photon
`
`detector, and which provide a region of intersection between a field of emission of
`
`the optical emitter and a field of detection of the photon detector. Id., 53-61, FIGS.
`
`5-7. The floor surface normally occupies the region of intersection of the sensor
`
`subsystem, until the robot approaches stairs or a drop that remove the floor surface
`
`from view of the sensor and causes the robot to “simply redirect.” Id., 6:59-64.
`
`
`
`2
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`The inventors identified a number of advantages of their techniques in the
`
`’308 patent, such as “provid[ing] a sensor subsystem for a robot which consumes a
`
`minimal amount of power,” “provid[ing] a sensor subsystem which is unaffected
`
`by surfaces of different reflectivity,” and providing a robot that can autonomously
`
`clean a room in “a wall following mode” and “a random bounce mode.” E.g., id.,
`
`1:51-2:19, 5:31-43, 6:56-58.
`
`III. LEVEL OF ORDINARY SKILL IN THE ART
`Since the instant Petition was filed, the ITC in a co-pending case (Inv. No.
`
`337-TA-1057) issued a claim construction order (IR2003) that addressed the
`
`definition of a person of ordinary skill in the art proposed by Petitioner in the ITC.
`
`In that order, the ITC agreed with Petitioner that experience in robotics was
`
`needed, but found that even more experience was needed than proposed by
`
`Petitioner. Indeed, the ITC found that a POSITA should have at least three years
`
`of experience in the design and implementation of robotics and embedded systems,
`
`rather than the one year of experience that Petitioner proposed in that proceeding.
`
`IR2003, p. 13. Patent Owner proposes that the Board adopt the same definition of
`
`a POSITA as the ITC; namely, “a person having an ordinary level of skill would
`
`hold a bachelor’s degree in physics, electrical engineering, mechanical
`
`engineering, computer science, or a related discipline, and have at least three years
`
`
`
`3
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`of experience in the design and implementation of robotics and embedded systems,
`
`or some other equivalent combination of education and experience.” Id.
`
`In contrast, Petitioner proposes here a definition of a POSITA that differs
`
`not only from the definition adopted in the ITC action, but also differs from
`
`Petitioner’s own definition that it proposed in the ITC. Petitioner provides no
`
`explanation why the definition here should differ. The inconsistent defintitions are
`
`shown below:
`
`Definition from IPR Petition
`
`Definition from ITC Proceeding
`
`“A POSITA in the field of the ’308
`
`A POSITA in the field of the ’308
`
`patent at the time of the earliest possible
`
`patent would have “a bachelor’s degree
`
`priority date (January 24, 2000) would
`
`in physics, electrical engineering,
`
`have had at least a Bachelor’s Degree in
`
`mechanical engineering, computer
`
`electrical engineering or computer
`
`science, or a related discipline, and have
`
`science, with an emphasis on
`
`at least one year of experience in the
`
`microprocessor based systems, and
`
`design and implementation of robotics
`
`between three and five years of
`
`and embedded systems, or some other
`
`experience in a practical setting, or
`
`equivalent combination of education and
`
`equivalent combinations of education
`
`experience.” (IR2003, p. 12)
`
`and experience.” (Petition, pp. 5-6)
`
`
`4
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Notably, the definition of a POSITA proposed by Petitioner in this IPR is
`
`devoid of any experience in mechanical design, physics, or even robotics. This
`
`definition seems aligned with the electrical and computer engineering background
`
`of Petitioner’s declarant, Dr. Neuhauser, whose CV describes his expertise to be in
`
`“the field of computer architecture and system design.” Ex. 1003, pp. 8-10,
`
`Appendix A. Although Dr. Neuhauser purports to have “analyzed … robotic
`
`manufacturing systems” in his capacity as a technical consultant for litigation
`
`matters, and “led teams” in the analysis of “robotic systems,” he does not mention
`
`any direct experience in robotics design, mechanical design, or physics, which are
`
`at issue in the challenged claims. Petition, p. 10.
`
`IV. CLAIM CONSTRUCTION
`The ’308 patent is unexpired, and therefore each claim is to be “given its
`
`broadest reasonable construction in light of the specification of the patent in which
`
`it appears” (“BRI”). 37 C.F.R. § 42.100(b). Federal Circuit “cases on BRI make
`
`clear that the proper BRI construction is not just the broadest construction, but
`
`rather the broadest reasonable construction in light of the specification.” In re Man
`
`Mach. Interface Techs. LLC, 822 F.3d 1282, 1287 (Fed. Cir. 2016).
`
`Patent Owner submits that the term re-direct of the ’308 patent requires
`
`additional consideration in view of Petitioner’s arguments. Patent Owner does not
`
`
`
`5
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`construe any other terms at this time, but reserves the right to do so at a later point
`
`in this proceeding.
`
`A.
`“re-direct[ing]” (claims 1 and 19)
`Independent claim 1 recites “a circuit in communication with the detector
`
`providing an output when an object is not present in the region thereby re-directing
`
`the autonomous robot.”2 Independent claim 19 recites “a circuit in communication
`
`with the detectors to re-direct the autonomous robot when the surface is not
`
`present in at least one region.”
`
`Petitioner does not provide an explicit construction of the terms “re-direct”
`
`or “re-directing.” Petition, 6. Instead, Petitioner offers an implicit, self-serving
`
`construction arguing that “re-directing” is met by “stopping.” Petition, 26-29.
`
`Petitioner’s implicit construction, however, is unreasonableit is inconsistent with
`
`the claims and the entire disclosure of the ’308 patent.
`
`As demonstrated below, in the context of the ’308 patent, the broadest
`
`reasonable interpretation of the terms “re-direct” or “re-directing” is “changing a
`
`direction of travel.” This construction is consistent with the intrinsic record, and
`
`supported by extrinsic evidence. Petitioner’s implicit construction is not.
`
`
`
`2 All emphasis added unless otherwise noted.
`
`6
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`1.
`The Intrinsic Record Confirms that “re-direct[ing]” Should be
`Construed as “changing a direction of travel”
`The intrinsic record fully supports Patent Owner’s construction that “re-
`
`direct[ing]” means “changing a direction of travel.” Petitioner’s overly broad
`
`construction equating “re-direct[ing]” with “stopping,” in contrast, finds no support
`
`in the intrinsic record.
`
`First, Patent Owner’s construction is supported by the language of the claims
`
`themselves. Claim 1 provides that the robot “rides on a surface,” and claim 19
`
`refers to “a surface upon which the autonomous robot travels.” Ex. 1001, 12:61-
`
`62, 14:10-11. With this language, claims 1 and 19 contemplate a robot’s riding or
`
`traveling on a surface. In this context, “re-directing” the robot involves changing
`
`the direction in which the robot is riding or traveling. Indeed, the terms “re-direct”
`
`and “re-directing” denote, from the stem “direct” and prefix “re,” that the robot
`
`changes direction.
`
`Additionally, the specification supports an interpretation of “re-directing” as
`
`changing a direction of travel. The ’308 patent’s autonomous robot has two modes
`
`of operation: a “random bounce mode” and a “wall following” mode. Ex. 1001,
`
`1:29-41, 5:20-22. In every disclosed example for both modes, the specification
`
`uses “re-direct” to indicate a change in the robot’s direction of travel.
`
`
`
`7
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`In the random bounce mode, upon encountering an obstacle (e.g., a step), the
`
`robot consistently changes direction and continues travel. For instance, the robot
`
`“move[s] in a straight line until the robot comes into contact with an obstacle” and
`
`“the robot then turns away from the obstacle and heads in a random direction.”
`
`Id., 1:30-35; see also 5:39-43, 2:12-14 (“If the floor does not occupy the
`
`predefined region, a stair or some other obstacle is present and the robot is directed
`
`away accordingly.”), Abstract (“A circuit in communication with a detector
`
`redirects the robot when the surface does not occupy the region to avoid
`
`obstacles.”), 2:38-40, 6:59-64, 7:37-39. Thus, upon detecting an obstacle in
`
`random bounce mode, the robot re-directs in a different direction. It never
`
`maintains its direction and stops.
`
`Similarly, in the wall following mode, the ’308 patent consistently describes
`
`redirection as involving continued travel after a change in direction. In that mode,
`
`“the robot encounters a wall, follows it for a time, and then returns to the random
`
`mode.” Id., 1:35-37. To remain in proximity of the wall during wall following,
`
`the robot is alternately “redirected” toward or away from the wall depending on
`
`whether the wall is presently detected by a wall sensor on the robot. Id., 3:7-10
`
`(“The circuit includes logic which redirects the robot away from the wall when the
`
`wall occupies the region and back towards the wall when the wall no longer
`
`occupies the region of intersection …”); see also 5:66-6:2, 7:1-3, 8:24:33, FIG. 17.
`8
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Consistent with this description, claims 4 and 5 of the ’308 patent use the
`
`term “redirects” in claiming features related to wall-following. Id., 13:10:15-18.
`
`For example, claim 4 recites “logic which redirects the robot away from the wall
`
`when the wall occupies the region and back towards the wall when the wall no
`
`longer occupies the region.” Id. With this usage, the claims align with the
`
`specification and confirm that, when claiming redirection, the ’308 patent means
`
`changing a direction of travel.
`
`Notably, the ’308 patent never discusses “stopping” the robot upon
`
`encountering an obstacle, and there is no example in the ’308 patent of the robot
`
`“re-directing” by “stopping.” Rather, as discussed above, every instance of re-
`
`direction described in the ’308 patent involves the robot continuing travel in a
`
`different direction from which the robot previously traveled. In fact, the
`
`specification twice uses “stop” in the context of stopping the robot’s turning, but
`
`not in the context of redirecting or stopping the robot’s travel. See Ex. 1001,
`
`10:33-35 (“[T]he robot monitors its wall sensor and if it detects a wall and then the
`
`wall detection goes away, the root [sic] stops turning.”); 10:37-39 (“If the robot has
`
`not seen its wall detector go on and then off by the time it reaches its maximum
`
`angle, it stops anyway.”).
`
`The only justification the Petition provides for its construction is an
`
`uncorroborated statement from Dr. Neuhauser that “[o]ne of ordinary skill would
`9
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`understand that the simple action of stopping the vehicle of Drunk-106 is
`
`‘redirecting’ it because it will no longer continue on the previous path.” Petition,
`
`p. 29 (citing Ex. 1003, ¶120). But Dr. Neuhauser’s statement is conclusory, and
`
`fails to address any evidence that would support his opinion. This type of
`
`conclusory analysis is entitled to no weight. See Zimmer Biomet Holdings, Inc., v.
`
`Four Mile Bay, LLC, IPR2016-00011, Paper 8 (Decision Denying Institution), p.
`
`11 (PTAB 2016) (citing Verlander v. Garner, 348 F.3d 1335, 1371 (Fed. Cir.
`
`2003)).
`
`Accordingly, the intrinsic record supports construing “re-direct” and “re-
`
`directing” to mean “changing a direction of travel.”
`
`2.
`Extrinsic Evidence Supports Patent Owner’s Construction
`Evidence outside the four corners of the ’308 patent also supports this
`
`construction. Literature in the field of robotics consistently refers to the term
`
`“redirect” (or variants of the term) to describe a change in the direction of travel.
`
`For instance, a research paper discussing autonomous robots from the mid-2000s
`
`described control logic that used “redirection values” to express an amount that an
`
`autonomous robot should change direction during travel for collision avoidance
`
`purposes. IR2004, p. 4. Patent literature also confirms that redirecting a robot
`
`involves changing its direction. See IR2006, 4:20-30, 5:43-49 (patent describing a
`
`robot that “redirect[s]” from a “forward direction of travel” to a reverse direction
`
`
`
`10
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`of travel that “cause[s] it to back up”); IR2005, ¶¶0005, 0009, 0012, Abstract
`
`(patent describing an autonomous cleaning robot that redirects by changing a
`
`direction of travel to stay within active and passively defined boundaries). Finally,
`
`dictionary definitions from the time of the ’308 patent further indicate that the
`
`plain meaning of “re-direct[ing]” involves a change in direction. See, e.g., IR2001,
`
`1464 (defining “re-direct” as “to change the direction or course of”); IR2002, 979
`
`(“[T]o change the course or direction of.”).
`
`In sum, Patent Owner’s construction is supported by the intrinsic and
`
`extrinsic evidence. Petitioner fails to address either. Accordingly, the Board
`
`should decline the Petition’s implicit construction and, instead, construe the terms
`
`“re-direct” and “re-directing” to mean “changing a direction of travel.”
`
`V. GROUNDS 1-4 SHOULD BE DENIED IN FULL BECAUSE THE
`CITED REFERENCES FAIL TO TEACH OR SUGGEST “RE-
`DIRECT[ING] THE AUTONOMOUS ROBOT”
`As discussed at Section IV.A, claim 1 recites “a circuit in communication
`
`with the detector providing an output when an object is not present in the region
`
`thereby re-directing the autonomous robot,” and claim 19 recites “a circuit in
`
`communication with the detectors to re-direct the autonomous robot when the
`
`surface is not present in at least one region.” The Petition’s reliance on Drunk-106
`
`for these features is misplaced.
`
`
`
`11
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Specifically, the Petition presents two theories for how Drunk-106 teaches
`
`the claimed “re-direct[ing],” but neither is availing. First, the Petition points to the
`
`following portion of Drunk-106 and wrongly asserts that “stopping” constitutes
`
`“re-directing” the robot:
`
`These infrared sensors ascertain the distance from the
`floor by means of their downward beaming infrared
`radiation and the reception of the reflected energy. If
`there is an increase in this distance beyond a prescribed
`limit, for example because one of the infrared sensors
`100, 101, 102, 103 is over a lower step while wheels 20,
`21, 22 of cleaning vehicle 1 are still supported on the
`floor above the step, the excessive difference in distance
`detected by the infrared sensor releases a signal which
`stops the operation of cleaning vehicle 1 and, under
`circumstances, releases an acoustical or optical signal to
`inform an operator. In this manner, it is ensured that the
`cleaning vehicle does not fall.
`
`Ex. 1004, 11:52-65 (cited by Petition, pp. 27-28).
`
`As discussed in Section IV.A, supra, the broadest reasonable interpretation
`
`of the terms “re-direct” and “re-directing” requires changing a direction of the
`
`robot’s travel. Drunk-106’s robot fails to meet this basic requirement, because it is
`
`configured to stoprather than change a direction of travelin response to
`
`detecting a cliff with its infrared sensors 100-103. See Ex. 1004, 11:52-65, 6:4-13.
`12
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Indeed, not only does Drunk-106 fail to describe changing directions, it fails to
`
`describe that the robot continues travel at all after detecting a cliff. Because
`
`Drunk-106’s stopping does not fall within the broadest reasonable interpretation of
`
`“re-direct[ing]” for the reasons discussed in Section IV.A, supra, the Petition’s
`
`first theory fails.
`
`The Petition also advances a second, back-up theory for how Drunk-106
`
`meets the “re-directing” features of claims 1 and 19. For the second theory, the
`
`Petition contends that “Drunk-106 also discloses that instead of stopping
`
`navigation upon cliff sensing by the downward sensors, the autonomous robot can
`
`be programmed to circumvent the ledge or cliff.” Petition, p. 28. This is simply
`
`untrueDrunk-106 never discloses circumventing a ledge or cliff.
`
`Specifically, Drunk-106 mentions cliff detection in only two passages. The
`
`first passage, at column 6, lines 3-13, describes “drop monitoring sensors” that are
`
`“aimed in a vertical direction” to detect a “step,” thereby causing the vehicle to
`
`“stop.” Ex. 1001, 6:3-13. The second passage, at column 11, lines 47-65,
`
`describes infrared sensors 100-103 detecting “a lower step” based on “the distance
`
`from the floor” increasing “beyond a prescribed limit,” which results in the release
`
`of “a signal which stops the operation of cleaning vehicle 1.” Id., 11:55-62. Both
`
`passages describe the robot stopping upon detecting a step or cliff. Neither
`
`
`
`13
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`passage, however, mentions circumventing the steps by changing a direction of
`
`travel.
`
`Faced without disclosure of step circumvention in Drunk-106, the Petition
`
`wrongly attempts to connect Drunk-106’s discussion of “circumvent[ing]” an
`
`above-ground obstacle using certain ultrasonic or optical “obstacle sensors” with
`
`the separate discussion of step detection using “drop monitoring sensors,” such as
`
`“infrared sensors 100-103.” Petition, pp. 28-29. This connection, however, is
`
`missing from Drunk-106. Although Drunk-106 mentions that “[o]bstacles may …
`
`take the shape of steps down which the vehicle might fall,” Drunk-106 never
`
`mentions that “steps” are detected using its “obstacle sensors,” or that the robot
`
`circumvents “steps” like any other above-ground obstacle. Petition, p. 28 (citing
`
`Ex. 1004, 6:3-4). To the contrary, as discussed above, Drunk-106 only mentions
`
`step detection using drop monitoring or infrared sensors, and only describes
`
`stopping and waiting for operator help when it encounters a step. See Ex. 1004,
`
`11:52-65. To the extent Petitioner contends that circumventing a step would be an
`
`obvious modification of Drunk’s vehicle, the Petition fails to provide any evidence
`
`or explanation for why a POSITA would have been led to such modification.
`
`Moreover, Drunk-106 itself explains why stopping, not re-directing, makes
`
`sense when Drunk-106’s robot detects a step. Specifically, Drunk-106’s robot
`
`does not navigate in the same way as the ’308 patent’s robot, which independently
`14
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`determines where to travel and in which mode to travel. Ex. 1001, 1:15-41, 5:18-
`
`43, FIGS. 4, 16-17, 22.
`
`In contrast, Drunk-106’s robot follows “a predetermined planned path” that
`
`is generated by “an off-line programming system.” Ex. 1004, 3:22-23, 3:49-51.
`
`Drunk-106’s robot follows the planned path by using sensor measurements to
`
`determine its distance to known markers along the path. Id., 4:9-17. Although
`
`Drunk-106’s robot is capable of making small deviations from the planned path to
`
`avoid objects “which were not there when the planned path was entered,” these
`
`types of objects are quite different from a set of steps, which are fixed in an
`
`environment and would not unexpectedly move into the robot’s path. Id., 5:60-6:2,
`
`11:66-12:3, 4:9-7, 5:60-6:2. If Drunk-106’s robot encounters a set of steps,
`
`something has gone significantly wrong.
`
`Additionally, steps present a dangerous hazard to the robot because of the
`
`risk of falling. Given the potential risk of falling and the unusual circumstance of
`
`encountering steps in navigating a planned path, as Drunk-106 describes multiple
`
`times, the appropriate action for the robot is to “stop” and summon a human
`
`operator. See id., 4:31-37, 6:55-65, 11:47-65. Stopping “ensure[s] that the
`
`cleaning vehicle does not fall.” Id., 11:63-65. Thus, in light of Drunk-106’s
`
`disclosure of stopping to achieve Drunk-106’s stated goal of ensuring that the
`
`
`
`15
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`robot does not fall, a POSITA would have been dissuaded from altering Drunk-
`
`106’s response to detection of steps.
`
`For these reasons, Petitioner has failed to demonstrate a reasonable
`
`likelihood of prevailing against independent claims 1 or 19 for any of Grounds 1-4.
`
`VI. THE PETITION’S ANALYSIS OF CLAIM 19 IS INCOMPLETE
`The Petition’s analysis of independent claim 19 in both Grounds 1 and 3
`
`fails to account for material differences between the features recited in claim 19
`
`and those recited in claim 1. Accordingly, Petitioner has not met its burden of
`
`showing a reasonable likelihood of claim 19 being unpatentable.
`
`A. THE PETITION’S ANALYSIS OF ELEMENTS 19(A) – (B)
`First, the Petition fails to establish that a POSITA would have incorporated
`
`multiple sensors of the types described in Jones-MR (Ground 1) or Everett
`
`(Ground 3) into Drunk-106’s robot. Elements 19a-19b of claim 19 recite:
`
` at least two emitters, each for emitting a beam having a
`field of emission toward a surface upon which the
`autonomous robot travels;
` at least two photon detectors, each having a field of
`view which intersects at least one field of emission at a
`region
`The Petition provides limited analysis for these elements of claim 19, arguing only
`
`that they are obvious for largely the same reasons as elements from claim 1. See
`
`Petition, pp. 37-39, 62-64. Although the Petition rightly acknowledges that claim
`16
`
`
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`19 differs from claim 1 in that claim 1 “does not expressly require multiple
`
`emitters” (or detectors) (id., pp. 38-39), the Petition fails to show that a POSITA
`
`would have been led to modify Drunk-106’s robot based on the teachings of either
`
`Jones-MR or Everett to meet the requirement of multiple emitters and detectors.
`
`For example, without evidence or further support, the Petition asserts that
`
`“[a] POSITA would have understood in view of Jones-MR that multiple GP2D02
`
`sensors would have been used on the Drunk-106 robot,” and, “[i]n particular, the
`
`robot would have used multiple GP2D02 sensors as the downward-facing cliff
`
`sensors (101) and (102) that are expressly disclosed in Drunk-106.” Petition, p.
`
`38.3 But these assertions are mere attorney argument, unsupported by the record.
`
`The Petition’s entire analysis of claim 19 in both Grounds 1 and 3 lacks citations to
`
`Dr. Neuhauser’s declaration, or any other exhibits. Without evidence, the Petition
`
`cannot meet its burden of proving unpatentability.
`
`
`
`3 The Petition provides a similar conclusory analysis for Ground 3 based on
`
`Drunk-106 in view of Everett. Petition, p. 63 (“A POSITA would therefore have
`
`understood that multiple convergent mode sensors of Everett would be used on the
`
`mobile robot of Drunk-106, including as the downward facing cliff sensors (101)
`
`and (102).”)
`
`
`
`17
`
`
`
`Case IPR2017-02078
`Attorney Docket No: 44360-0002IP1
`Indeed, the Petition’s argument that a POSITA would have replaced each of
`
`Drunk-106’s cliff sensors 101 and 102 with a respective sensor from Jones-MR or
`
`Everett, simply because Drunk-106 already discloses the use of multiple sensors, is
`
`conclusory and fails to address any number of practical issues that a POSITA
`
`would have considered before making the modification. For example, the Petition
`
`cites no evidence that addresses (1) whether multiple sensors from Jones-MR or
`
`Everett would be necessary in view of their particular capabilities and
`
`configurations, (2) whether implementing multiple sensors from Jones-MR or
`
`Everett would be feasible in vie