`Tel: 571-272-7822
`
`Paper No. 8
`Entered: March 12, 2018
`
`
`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 IPR2017-02061
`Patent 6,809,490 B2
`____________
`
`
`
`Before WILLIAM V. SAINDON, TERRENCE W. MCMILLIN, and
`AMANDA F. WIEKER, Administrative Patent Judges.
`
`SAINDON, Administrative Patent Judge.
`
`
`
`DECISION
`Granting Partial Institution of Inter Partes Review
`37 C.F.R. § 42.108
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`
`INTRODUCTION
`I.
`Shenzhen Zhiyi Technology Co. Ltd., d/b/a iLife, (“Petitioner”) filed
`a Petition requesting an inter partes review of claims 1–3, 7, 12, and 42 of
`U.S. Patent No. 6,809,490 B2 (Ex. 1001, “the ’490 patent”). Paper 1
`(“Pet.”). iRobot Corp. (“Patent Owner”) filed a Preliminary Response.
`Paper 7 (“Prelim. Resp.”).
`We have authority under 35 U.S.C. § 314, which provides that an
`inter partes review may not be instituted unless the information presented in
`the Petition and the Preliminary Response shows that “there is a reasonable
`likelihood that the petitioner would prevail with respect to at least 1 of the
`claims challenged in the petition.” 35 U.S.C. § 314; see also 37 C.F.R
`§ 42.4(a) (“The Board institutes the trial on behalf of the Director.”). Taking
`into account the arguments presented in the Preliminary Response, we
`conclude that the information presented in the Petition does not establish a
`reasonable likelihood that Petitioner would prevail with respect to the claims
`challenged on the basis of anticipation (claims 1–3, 7, and 12) but does
`establish a reasonable likelihood that Petitioner would prevail on the basis of
`obviousness (claim 42). Accordingly, we institute an inter partes review for
`claim 42 only.
`
`A. Related Matters
`The parties identify the following matters related to the ’490 patent
`(Pet. 1; Paper 4, 2–3; Paper 6, 2–3):
`In re Certain Robotic Vacuum Cleaning Devices And Components
`Thereof Such As Spare Parts, Inv. No. 337-TA-1057 (USITC);
`iRobot Corp. v. Hoover, Inc., Case No. 1:17-cv-10647 (D. Mass.);
`
` 2
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`iRobot Corp. v. Bissell Homecare, Inc., Case No. 1:17-cv-10649 (D.
`Mass.);
`iRobot Corp. v. Bobsweep, Inc., Case No. 1:17-cv-10651 (D. Mass.);
`
`and
`
`iRobot Corp. v. Shenzhen Zhiyi Technology Co. d/b/a iLife, Case No.
`1:17-cv-10652 (D. Mass.).
`In addition, Petitioner has filed petitions challenging five of Patent
`Owner’s other patents in IPR2017-02050 (U.S. Patent No. 9,038,233 B2),
`IPR2017-02078 (U.S. Patent No. 7,155,308 B2), IPR2017-02133
`(U.S. Patent No. 8,600,553 B2), IPR2017-02137 (U.S. Patent No. 9,486,924
`B2), and IPR2018-00005 (U.S. Patent No. 8,474,090 B2).
`
`B. The ’490 Patent
`The ’490 patent is directed to a mobile robot used, e.g., in vacuum
`cleaning or mowing. Ex. 1001, 1:9–12. A challenge acknowledged in the
`prior art is designing an algorithm that allows the robot to cover all of an
`area of unknown geometry in an efficient amount of time. Id. at 1:22–2:19.
`The ’490 patent discloses a robot that moves through various operational
`modes, including spot cleaning, edge cleaning, and room cleaning modes to
`effectively cover the area. Id. at 8:35–47. These high-level operating modes
`are, in turn, effected by the robot giving priority to various behaviors
`dictating how the robot reacts in various situations (e.g., hitting a wall). Id.
`at 8:48–9:5, 13:26–35.
`
`C. Challenged Claims
`Petitioner challenges claims 1–3, 7, 12, and 42 of the ’490 patent.
`Independent claims 1 and 42 are reproduced below.
`
` 3
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`1. A mobile robot comprising:
`(a) means for moving the robot over a surface;
`(b) an obstacle detection sensor;
`(c) and a control system operatively connected to said obstacle
`detection sensor and said means for moving;
`(d) said control system configured to operate the robot in a
`plurality of operational modes and to select from among the
`plurality of modes in real time in response to signals
`generated by the obstacle detection sensor, said plurality of
`operational modes comprising: a spot-coverage mode
`whereby the robot operates in an isolated area, an obstacle
`following mode whereby said robot travels adjacent to an
`obstacle, and a bounce mode whereby the robot travels
`substantially in a direction away from an obstacle after
`encountering the obstacle, and wherein, when in the obstacle
`following mode, the robot travels adjacent to an obstacle for
`a distance at least twice the work width of the robot.
`
`
`42. A mobile robot comprising:
`(a) means for moving the robot over a surface;
`(b) an obstacle detection sensor;
`(c) a cliff sensor; and
`(d) a control system operatively connected to said obstacle
`detection sensor, said cliff sensor, and said means for moving;
`(e) said control system configured to operate the robot in a
`plurality of operational modes, said plurality of operational
`modes comprising: a spot-coverage mode whereby the robot
`operates in an isolated area, an obstacle following mode
`whereby said robot travels adjacent to an obstacle for a
`distance at least twice the work width of the robot, and a
`bounce mode whereby the robot travels substantially in a
`direction away from an obstacle after encountering the
`obstacle.
`
`
`
` 4
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`
`D. Prior Art and Asserted Grounds
`Petitioner asserts that claims 1–3, 7, and 12 are anticipated by Ueno-
`642,1 and that claim 42 would have been obvious in view of Ueno-642 and
`Bissett-612.2 Pet. 4.
`
`II. PATENTABILITY ANALYSIS
`A. Claim Construction
`No terms require construction for purposes of this decision. See Vivid
`Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)
`(only those terms that are in controversy need to be construed, and only to
`the extent necessary to resolve the controversy).
`
`B. Anticipation by Ueno-642
`(Claims 1, 2, 3, 7, and 12)
`Petitioner asserts that Ueno-642 anticipates independent claim 1
`(Pet. 14–32) and claims 2, 3, 7, and 12, which depend therefrom (id. at 32–
`35). As to claim 1, Petitioner asserts that Ueno-642 discloses a robot having
`wheels 3, 4 for moving the robot over the surface. Id. at 15–16 (citing, e.g.,
`Ex. 1004 ¶ 15). Petitioner asserts that the robot in Ueno-642 has an obstacle
`detection sensor and a control system. Id. at 17–18 (citing, e.g., Ex. 1004
`¶¶ 16, 21). For element 1(d), Petitioner asserts that Ueno-642 discloses a
`control system having a plurality of operational modes (spiral, border-
`following, and random) and can switch among them “in real time in
`
`
`1 Japanese Unexamined Patent Application Publication No. H11-212642,
`pub. Aug. 6, 1999 (Ex. 1004).
`2 U.S. Patent No. 6,493,612 B1, iss. Dec. 10, 2002 (Ex. 1005).
`
` 5
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`response to signals generated by the obstacle detection sensor.” Id. at 18–22
`(citing, e.g., Ex. 1004 ¶¶ 23–26, 30, 33, 35, 38–40). Petitioner then goes
`into more detail regarding the various operational modes in Ueno-642. Id. at
`2–32.
`
`Patent Owner focuses its Preliminary Response on element 1(d).
`Prelim. Resp. 23–29. Patent Owner argues that Ueno-642 shows various
`modes of operation, but not selection between the modes “in response to
`signals generated by the obstacle detection sensor,” as required by claim 1.
`Id. at 24.
`We have reviewed Petitioner’s and Patent Owner’s arguments and
`evidence with respect to element 1(d), and are persuaded that Petitioner has
`not shown sufficiently that Ueno-642 describes this claim element. See
`Karsten Mfg. Corp. v. Cleveland Golf Co., 242 F.3d 1376, 1383 (Fed. Cir.
`2001) (stating that, to establish anticipation, “all of the elements and
`limitations of the claim must be shown in a single prior reference, arranged
`as in the claim”). Our reasons follow.
`The critical language in claim 1 is that the control system “select[s]
`from among the plurality of modes . . . in response to signals generated by
`the obstacle detection sensor” (emphasis added). This requires, in practical
`application, that the system can choose a mode in which to operate (“select
`from among”), based on inputs from the obstacle sensor (“in response to
`signals”).
`The first mode in Ueno-642 identified by Petitioner is the spiral mode.
`Pet. 19 (“[I]n Ueno-642, the robot operates in at least three modes: ‘spiral
`travel,’ ‘border-following travel’ and ‘random travel.’”). Petitioner directs
`us to a disclosure in Ueno-642 where, “[i]f a sensor 26 or 25L detects a
`
` 6
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`boundary, a wall surface or an obstacle,” the spiral is determined to be
`complete and a stop instruction is given. Pet. 20 (quoting Ex. 1004 ¶ 40);
`see also Ex. 1004, Fig. 10 (depicting the spiral mode algorithm), Fig. 9
`(depicting the overall algorithm, where, when spiral mode is determined to
`be finished at S59, the algorithm goes back to 1 and then S1 where the travel
`mode pointer is updated and read, i.e., the next mode begins). Notably,
`however, signals from the obstacle sensors in the spiral mode determine
`when to exit the spiral mode. Determining when to exit the present mode is
`not the same as selecting the next mode based upon signals generated by the
`obstacle detection sensor.
`As Patent Owner points out, the three modes in Ueno-642 are chosen
`based on the status of a “travel mode pointer.” Prelim. Resp. 26–27 (citing
`Ex. 1004 ¶¶ 35, 37). We do not find this alone to preclude that the travel
`mode pointer was set based on obstacle sensor input. However, Petitioner
`does not explain or direct us to any discussion in Ueno-642 that sufficiently
`tells us which travel mode pointer value is selected in response to obstacle
`sensor input. At best, we have a statement that Figure 9 explains how “the
`robot travel control operations [are] based on the output signal of each
`sensor described above.” Ex. 1004 ¶ 35 (citing Pet. 20). The rest of that
`paragraph, however, describes how the travel pointer value is not selected
`from among its possible values but rather merely moves to the next part of a
`pre-ordained, repeated sequence:
`As to in what sequence these travel modes are executed varies
`depending on the size and shape of the region planned for travel
`and also if there are obstacles, but the inventors involved herein
`confirmed that a good result later described was obtained by
`simulations in which the combination of a spiral travel, a random
`
` 7
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`travel and a border-following travel and a random travel are
`executed repeatedly in this sequence.
`Of course, it is possible that by making other various
`combination sequences, for instance, spiral - border-following -
`random as a set, these can be repeated by the same sequence or
`changed sequence, but as in spiral - random - border-following -
`random - spiral, at least one of border-following travel and spiral
`travel is alternatively executed between two random travels that
`are before and after. And it can be designed such that the worker
`can set up these and register each time or can preregister and
`select and set up at work start time. Like this, the combination
`sequence of the travel modes thus set up and registered is
`temporarily stored in memory and the travel mode to be executed
`as of now is instructed sequentially by aforementioned pointer
`(not shown in figure).
`Ex. 1004 ¶¶ 35–36 (emphases added).
`
`In our view, these paragraphs characterize the mode selection in
`Ueno-642 as a pre-ordained sequence of modes rather than a selection based
`on obstacle sensor input. See also id. ¶ 35 (noting that the travel mode
`pointer starts at 0 and goes to 1 for the initial mode, indicating a mode
`scheme such as 1=spiral, 2=random, 3=border, which could be repeated to
`provide the sequence described)
`To the extent Petitioner’s argument is that exiting the spiral mode due
`to obstacle detection effectively selects the next mode in sequence, we
`determine that is not what the claims require. Instead, the claims recite that
`the system “selects from among the plurality of modes” (emphasis added),
`and we understand the plain meaning of this phrase to be that the control
`system has several options to choose from, and that choice is in response to,
`i.e., is based on, input from the sensor. In Ueno-642, there does not appear
`to be a choice made “among the plurality of modes” by the control system,
`but rather it must simply move to the pre-ordained next mode. In contrast,
`
` 8
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`claim 1 requires the control system to choose the mode from among the
`claimed modes.
`Nevertheless, even if we were to be persuaded that exiting one mode
`is effectively choosing the next, albeit preprogrammed, mode, we agree with
`Patent Owner (Prelim. Resp. 27) that Petitioner has not shown how Ueno-
`642 discloses exiting the border-following mode using obstacle detection
`because it appears the border-following mode only exits upon an expiration
`of a timer or a crossing of a distance threshold. Ex. 1004 ¶¶ 23 (“The
`border-following travel pattern . . . is continued for a planned time.”), 25
`(“[T]he border-following travel is stopped after continuing for a planned
`time (or distance), it moves to a random travel mode.”), Fig. 4; Pet. 20
`(citing Exhibit 1004 ¶¶ 23–25). Thus, even under the alternative reading of
`“selecting . . . in response to” (which we determine is incorrect), Petitioner
`has not shown how Ueno-642 satisfies the limitations of the claim.
`We also agree with Patent Owner’s analysis as to the disclosures in
`Ueno-642 regarding priority of operations and action plans not being a
`disclosure of a control system selecting modes as required by claim 1. See,
`e.g., Ex. 1004 ¶ 21 (“[O]peration decision unit 18 . . . generates an action
`plan depending on the signal status from sensor 25L, 26, 5A and outputs an
`execution request.”). As explained in paragraphs 21 and 22, Ueno-642 is
`discussing how the robot handles the immediate consequences of hitting an
`object, for example, because the operations among which it selects are not
`the operating modes but rather smaller tasks such as retreat, advance, stop,
`rotate, etc. See id. ¶¶ 21 (“[A]ction plans that were individually generated
`based on information from each sensor are piled up and the entire robot
`operations, that is, the operations such as a straight advance, a retreat, a stop
`
` 9
`
`
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`and a slow rotation, a fast rotation and a pivot turn, an ultra-pivot turn etc[.]
`are decided.”), 22 (similar). The next paragraph discusses the operational
`modes, bringing into contrast the sensor-driven tasks of the prior paragraph.
`Id. ¶ 23. Thus, the action plan determines what the robot does within a
`given mode, but is not itself selection of a mode. For example, the action
`plan helps the robot decide whether to stop if it thinks it hits something, for
`example a wall in the random mode, and then helps it decide how to
`maneuver into a turn away from that wall. See, e.g., id. ¶¶ 42–44
`(explaining how the robot uses the action plan to maneuver when it hits a
`wall in the spiral mode). This is not a disclosure of selection among
`operating modes as required by the claim.
`Having reviewed the arguments and evidence before us, we determine
`that Petitioner has not demonstrated a reasonable likelihood of success of
`showing that claim 1 is anticipated by Ueno-642. Petitioner’s ground as to
`dependent claims 2, 3, 7, 12 is deficient for the same reason.
`
`C. Patent Owner’s 35 U.S.C. § 325(d) Argument
`Patent Owner argues that the Petition should be denied because it
`relies on “substantially the same prior art or arguments previously . . .
`presented to the Office.” Prelim. Resp. 11 (quoting 35 U.S.C. § 325(d)); see
`also id. at 11–23 (setting forth the full § 325(d) argument). We have denied
`the Petition as to the anticipation ground on the merits; there is no reason to
`analyze that ground in view of § 325(d). As to the obviousness ground, we
`determine that the Ueno-642/Bisset ground is not substantially the same
`prior art or argument previously presented to the Office, for the reasons set
`forth below, such that 35 U.S.C. § 325(d) is not a reason to deny the
`Petition.
`
`
`10
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`In the underlying prosecution of the application leading to the ’490
`patent, the examiner rejected the claims as obvious in view of Noonan3 and
`Ueno-025.4 Ex. 2001, 756–763 (Sept. 3, 2003 Non-Final Office Action),
`317–324 (Dec. 29, 2003 Final Office Action). In the Non-Final Action, the
`examiner found that Noonan discloses a spot coverage mode and an obstacle
`following mode, but not a bouncing mode. Id. at 759 (“[G]uide path [in
`Fig.] 1A represents a pattern for spot coverage mode. Fig. 1C represents a
`pattern that is required for obstacle following mode.”). The examiner found
`Ueno-025 teaches a spot coverage mode and a random, or bounce, mode (id.
`at 760), and found that the reason to combine was to “improve robot work
`efficiency” (id. at 761).
`The applicant amended the claims to specify, inter alia, that the
`modes were operational modes. Id. at 343, 345, 347. The applicant then
`pointed out that Noonan does not teach an obstacle-following mechanism
`but rather an obstacle-avoiding mechanism. Id. at 353. The applicant
`further pointed out that Noonan’s device merely follows a predetermined
`guided path, and modifying it to have a random bounce mode “would
`completely change the nature of Noonan’s device.” Id. at 362.
`The examiner indicated the applicant’s arguments were considered,
`and maintained the same rejection, but did not provide much further
`explanation. Id. at 319 (stating, “Applicant’s arguments have been
`considered. However, . . . the examiner[’]s responsibility is to interpret
`claim language as broad as possible [sic].”), 320 (effectively relying on the
`
`
`3 U.S. Patent No. 5,204,814, iss. Apr. 20, 1993.
`4 U.S. Patent No. 6,076,025, iss. June 13, 2000 (Ex. 2002).
`
`11
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`same rejection). On March 11, 2004, an interview occurred after the final
`rejection, wherein the examiner noted the interview involved discussion of
`potential new claim language specifying that “in obstacle following mode,
`the robot travels adjacent to an obstacle for a distance at least twice the work
`width of the robot,” and then indicated that a search would be provided. Id.
`at 67.
`
`For the majority of claims, the applicant subsequently added not only
`the “twice the work width” limitation (id. at 75, 80, 80–81; but cf. id. at 79
`(not adding the “twice” limitation to prosecution claim 36)), but also added
`some language not identified as having been discussed in the interview,
`directed to the control system “select[ing] from among the plurality of
`modes in real time in response to signals” (id. at 75, 76, 80; but cf. id. at 79
`(not amending prosecution claim 36 to require “selecting among,” but rather
`“alternat[ing] modes”)), i.e., the limitation we found missing in Ueno-642.
`The examiner then issued a notice of allowance, stating the reason for
`allowance is “based on provided amendment after final office action and
`further search.” Id. at 63–65. Reviewing the prosecution history, we find
`that the issues raised during prosecution largely involve Noonan, and to the
`extent they involve Ueno-025, the issue stems from the applicant’s position
`that Ueno-025 operated in a disparate fashion compared to Noonan in the
`combination.
`The prior art rejection faced by the applicant during examination is
`much different from the prior art ground offered by Petitioner. Ueno-642
`includes the obstacle-following limitations that the applicant pointed out
`were problematic in the examiner’s rejection using Noonan, and further
`these limitations were not alleged to have been present in Ueno-025. Ueno-
`
`
`12
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`025 and Ueno-642 do include some similar disclosures, as Patent Owner
`points out. See Prelim. Resp. 13–14 (showing Figures 5a–b are the same
`and Figures 6a–c are the same in each), 18–21 (showing various isolated
`passages share similar language). However, there is only a passing
`resemblance to the way Ueno-642 is used here compared to Ueno-025 in the
`rejection. For example, Ueno-642 is the primary reference whereas Ueno-
`025 was a secondary reference. Ueno-642 is used to teach each of the
`claimed modes in the same device, being combined with Bisset for a
`particular sensor, whereas the examiner was combining a preprogrammed
`path robot (Noonan) with a random path robot (Ueno-025) in the rejection.
`Ueno-642 allegedly teaches all three modes, whereas Ueno-025 does not.
`Noonan is not present in the ground, and Bisset’s role in the ground is
`secondary in nature, to teach a cliff sensor in a vacuum robot. On whole, the
`Ueno-642/Bisset prior art ground bears little resemblance to the
`Noonan/Ueno-025 prior art rejection.
`In conclusion, we find no substantial similarity between Petitioner’s
`ground (i.e., the prior art) and the examiner’s rejection (i.e., the prior art
`previously considered), and find no substantial similarity between the issues
`raised in Petitioner’s ground (e.g., Ueno-642’s specific teachings regarding
`three modes) and the issues raised during prosecution (e.g., Noonan’s
`specific teachings, combinability of predetermined path and random path
`robots). Because the Office has not considered substantially the same prior
`art and has not considered substantially the same arguments, we decline to
`deny the Petition on the basis of 35 U.S.C. § 325(d).
`
`
`13
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`D. Obviousness over Ueno-642 and Bisset
`(Claim 42)
`According to Petitioner, the main difference between claim 42 and
`claim 1 is that claim 42 requires a cliff sensor. Pet. 35. Petitioner’s ground
`relies on the same teachings in Ueno-642 for the claimed control system and
`operating modes (id. at 40–41), but claim 42, unlike claim 1, does not set
`forth a requirement that the modes are selected in any particular manner.
`The deficiency we identified with respect to Petitioner’s ground asserting
`anticipation of claim 1 is not found in this ground. As to the additional cliff
`sensor limitation, Petitioner asserts that Bisset describes a cliff sensor and
`that a person of ordinary skill in the art would have considered it obvious to
`add a cliff sensor in Ueno-642 in order to prevent the robot from falling off a
`cliff (e.g., stairs), as taught in Bisset. See Pet. 35–38 (citing, e.g., Ex. 1005,
`1:46–47, 7:51–8:4; Ex. 1003 ¶ 53); KSR Int’l Co. v. Teleflex, Inc., 550 U.S.
`398, 417 (2007) (“[I]f a technique has been used to improve one device, and
`a person of ordinary skill in the art would recognize that it would improve
`similar devices in the same way, using the technique is obvious unless its
`actual application is beyond his or her skill.”).
`Reviewing Petitioner’s arguments and evidence,5 we are persuaded
`that Petitioner has shown a reasonable likelihood of establishing the subject
`matter of claim 42 to have been obvious in view of Ueno-642 and Bisset.
`
`
`5 Patent Owner’s Preliminary Response does not discuss the merits of this
`ground.
`
`
`14
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`
`III. ORDER
`In view of the foregoing, it is hereby:
`ORDERED that an inter partes review of the ’490 patent is instituted,
`commencing on the entry date of this Order, and pursuant to 35 U.S.C.
`§ 314(c) and 37 C.F.R. § 42.4, notice is hereby given of the institution of a
`trial; and
`FURTHER ORDERD that the inter partes review is limited to
`determining whether claim 42 of the ’490 patent would have been obvious in
`light of the teachings of Ueno-642 and Bisset.
`
`
`
`
`
`
`
`15
`
`
`
`
`
`IPR2017-02061
`Patent 6,809,490 B2
`PETITIONER:
`
`Patrick McCarthy
`mccarthyp@gtlaw.com
`
`Cameron Nelson
`nelsonc@gtlaw.com
`
`
`PATENT OWNER:
`
`Walter Renner
`axf-ptab@fr.com
`
`Jeremy Monaldo
`jjm@fr.com
`
`Tonya Drake
`tdrake@irobot.com
`
`Patrick Bisenius
`bisenius@fr.com
`
`
`16
`
`
`
`