DAEDLUS BLUE LLC

Patent Trials and Appeals Board

Feb 3, 2022

IPR2020-01475 (P.T.A.B. Feb. 3, 2022)

Trials@uspto.gov Paper 31 571-272-7822 Date: February 3, 2022 UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD DJI EUROPE B.V., Petitioner, v. DAEDALUS BLUE LLC, Patent Owner. IPR2020-01475 Patent 7,228,232 B2 Before BARRY L. GROSSMAN, ERIC C. JESCHKE, and ARTHUR M. PESLAK, Administrative Patent Judges. PESLAK, Administrative Patent Judge. JUDGMENT Final Written Decision Determining No Challenged Claims Unpatentable 35 U.S.C. § 318(a) IPR2020-01475 Patent 7,228,232 B2 2 I. INTRODUCTION Petitioner, DJI Europe B.V., filed a Petition (Paper 2, “Pet.”) requesting an inter partes review of claims 7-12 (the “challenged claims”) of U.S. Patent No. 7,228,232 B2 (Ex. 1001, “the ’232 patent”). Petitioner filed a Declaration of Dr. R. John Hansman, Jr. in support of the Petition. Ex. 1003. Patent Owner, Daedalus Blue LLC, timely filed a Preliminary Response (Paper 7). Taking into account the arguments presented in Patent Owner’s Preliminary Response, we determined there was a reasonable likelihood Petitioner would prevail in its contention that at least one of the challenged claims of the ’232 patent is unpatentable under 35 U.S.C. § 103(a). On February 12, 2021, we instituted this inter partes review as to the challenged claims and all grounds presented in the Petition. Paper 14 (“Dec.”). During the course of trial, Patent Owner filed a response to the Petition. Paper 21 (“PO Resp.”). In support of the Patent Owner Response, Patent Owner filed a Declaration of Dr. Daniel S. Kaputa. Ex. 2011. Petitioner filed a Reply to the Patent Owner Response. Paper 23 (“Pet. Reply”). Patent Owner then filed a Sur-reply. Paper 24 (“Sur-reply”). An oral hearing was held on November 16, 2021, and a transcript of the hearing is entered into the record. Paper 30 (“Tr.”). We have jurisdiction under 35 U.S.C. § 6. This is a Final Written Decision under 35 U.S.C. § 318(a) as to the patentability of the challenged claims of the ’232 patent. For the reasons discussed below, we determine Petitioner does not establish by a preponderance of the evidence that any challenged claim of the ’232 patent is unpatentable. IPR2020-01475 Patent 7,228,232 B2 3 A. Related Matters The parties state that the ’232 patent is asserted in Daedalus Blue LLC v. SZ DJI Technology Co., Ltd., & DJI Europe B.V., Case No. 6:20-cv-0073 (W.D. Tex.). Pet. 91; Paper 5, 2. The ’232 patent is also the subject of IPR2020-01474, filed by Petitioner concurrently with this proceeding. Pet. 91. B. Real Parties-in-Interest Petitioner identifies itself and DJI Technology Inc., SZ DJI Technology Co., Ltd., iFlight Technology Company Limited, DJI Japan K.K., and DJI Research LLC as real parties-in-interest. Pet. 91. Patent Owner identifies itself as the only real party in interest. Paper 5, 2. C. The ’232 Patent (Ex. 1001) The ’232 patent is titled Navigating a UAV1 with Obstacle Avoidance Algorithms. Ex. 1001, code (54). The ’232 patent issued on June 5, 2007 from an application filed on January 24, 2005. Id. at codes (45), (22). The ’232 patent is directed to “[m]ethods, systems, and computer program products . . . for navigating a UAV that include piloting the UAV, under control of a navigation computer, in accordance with a navigation algorithm.” Ex. 1001, 1:34-37. Embodiments of the ’232 patent “include reading from the GPS receiver a sequence of GPS data, anticipating a future position of the UAV, identifying an obstacle in dependence upon the future position, selecting an obstacle avoidance algorithm, and piloting the UAV in accordance with an obstacle avoidance algorithm.” Id. at 1:37-42. Figure 1 of the ’232 patent, reproduced below, illustrates various components of a system for navigating an UAV. Ex. 1001, 4:48-50. 1 “UAV” is short hand for an “unmanned aerial vehicle.” Pet. 1. IPR2020-01475 Patent 7,228,232 B2 4 Figure 1 of the ’232 patent “illustrat[es] relations among components of an exemplary system for navigating a UAV.” Id. at 2:10-12. In the system shown in Figure 1, UAV 100 “includes a GPS (Global Positioning System) receiver (not shown) that receives a stream of GPS data from satellites (190, 192).” Ex. 1001, 4:50-53. UAV 100 is navigated “by IPR2020-01475 Patent 7,228,232 B2 5 receiving in a remote control device a user’s selection of a GUI [graphical user interface] map pixel that represents a waypoint for UAV navigation.” Id. at 4:57-60. Exemplary remote control devices include mobile telephone 110, laptop computer 106, and personal digital assistant 120. Id. at 4:61-64. The remote control devices “map the pixel[] location on the GUI to Earth coordinates of a waypoint and . . . transmit the coordinates of the waypoint to the UAV (100)” through wireless network 102. Id. at 5:8-13. The system shown in Figure 1 “is capable of operating a UAV to read a starting position from a GPS receiver . . . on the UAV and pilot the UAV, under control of a navigation computer on the UAV, from a starting position to a waypoint in accordance with a navigation algorithm.” Id. at 5:30-35. The system also reads “a sequence of GPS data, anticipat[es] a future position of the UAV, identif[ies] an obstacle in dependence upon the future position, select[s] an obstacle avoidance algorithm, and pilot[s] the UAV in accordance with an obstacle avoidance algorithm.” Id. at 6:43-49. D. Challenged Claims Claim 7 is the only challenged independent claim and is reproduced below: A system for navigating a UAV, the system comprising: means for piloting the UAV, under control of a navigation computer in accordance with a navigation algorithm; means for reading from a GPS receiver a sequence of GPS data; means for anticipating a future position of the UAV in dependence upon the sequence of GPS data; means for identifying an obstacle in dependence upon the future position; means for selecting an obstacle avoidance algorithm; and means for piloting the UAV in accordance with the selected obstacle avoidance algorithm. Ex. 1001, 21:60-22:3. IPR2020-01475 Patent 7,228,232 B2 6 E. Prior Art and Asserted Grounds Petitioner asserts that claims 7-12 would have been unpatentable on the following grounds: Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis 7, 12 103(a)2 Sanders3 8 103(a) Sanders, Pilley4 9 103(a) Sanders, Suiter5 10, 11 103(a) Sanders, Pappas6 7-9, 12 103(a) Duggan7, Pilley 10, 11 103(a) Duggan, Pilley, Sainthuile8 II. ANALYSIS A. Overview Petitioner bears the burden of establishing the unpatentability of the challenged claims by a preponderance of the evidence. 35 U.S.C. § 316(e) (2018); 37 C.F.R. § 42.1(d) (2020). This burden of persuasion never shifts to Patent Owner. Dynamic Drinkware, LLC v. Nat’l Graphics, Inc., 800 F.3d 1375, 1378 (Fed. Cir. 2015). 2 The Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29, 125 Stat. 284, 287-88 (2011), amended 35 U.S.C. § 102. Because the ’904 patent was filed before the effective date of the relevant amendment, the pre- AIA version of § 103 applies. 3 Christopher P. Sanders, Real-time Collision Avoidance for Autonomous Air Vehicles, (M.I.T. 1998) (Ex. 1005) (“Sanders”). 4 U.S. Patent No. 5,867,804 issued Feb. 2, 1999 (Ex. 1010) (“Pilley”). 5 U.S. Patent No. 6,690,299 B1 issued Feb.10, 2004 (Ex. 1011) (“Suiter”). 6 George J. Pappas, Conflict Resolution for Multi-Agent Hybrid Systems, (1996) (Ex. 1015) (“Pappas”). 7 U.S. Patent No. 7,343,232 B2 issued Mar. 11, 2008 (Ex. 1012) (“Duggan”). 8 U.S. Patent No. 6,546,338 B2 issued April 8, 2003 (Ex. 1013) (“Sainthuile”). IPR2020-01475 Patent 7,228,232 B2 7 A claim is unpatentable under § 103(a) if the differences between the claimed subject matter and the prior art are such that the subject matter, as a whole, would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406 (2007). The question of obviousness is resolved on the basis of underlying factual determinations, including (1) the scope and content of the prior art; (2) any differences between the claimed subject matter and the prior art; (3) the level of skill in the art; and (4) when in evidence, objective indicia of non-obviousness (i.e., secondary considerations).9 Graham v. John Deere Co., 383 U.S. 1, 17-18 (1966). B. Level of Ordinary Skill in the Art Petitioner contends that a skilled artisan “would have had a bachelor’s degree in aeronautical engineering, electrical engineering, computer science or equivalent training and experience, and at least two years of experience in the field of unmanned vehicles.” Pet. 7-8 (citing Ex. 1003 ¶ 20). Petitioner further contends that “[a]dditional work or research experience can substitute for less or different education, and vice-versa.” Id. at 8. Patent Owner appears to apply Petitioner’s level of ordinary skill in the art. See PO Resp. 17-18. Based on the complete record, we adopt Petitioner’s proposed level of ordinary skill in the art because it comports with the level of skill reflected in the ’232 patent and the prior art of record. 9 Neither party has submitted evidence concerning secondary considerations. IPR2020-01475 Patent 7,228,232 B2 8 C. Claim Construction We apply the same claim construction standard used by Article III federal courts and the International Trade Commission, both of which follow Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc), and its progeny. 37 C.F.R. § 42.100(b). Accordingly, we construe each challenged claim of the ’232 patent to generally be “the ordinary and customary meaning of such claim as understood by one of ordinary skill in the art and the prosecution history pertaining to the patent.” Id. Petitioner contends that no claim terms require express construction. Pet. 8. Petitioner submits a chart detailing disclosure in the ’232 patent of the corresponding function and structure for the means plus function claim elements recited in claims 7-12. Id. at 8-17. Claim 7 recites “means for selecting an obstacle avoidance algorithm.” Ex. 1001, 22:1. In our Decision on Institution, we explained “Patent Owner’s contentions are implicitly based on a construction of the phrase in claim 7 ‘means for selecting an obstacle avoidance algorithm’ as ‘means for selecting an obstacle avoidance algorithm from a plurality of collision avoidance algorithms.’” Dec. 25. Notwithstanding this, we explained the language of claim 7 does not appear to require selecting from more than one collision avoidance algorithm. Id. at 26. Further, we invited the parties to brief the proper construction of claim 7. Id. at 27. Patent Owner now explicitly advocates for this construction. PO Resp. 18-21. In particular, Patent Owner contends this term requires “a selection from multiple possible collision avoidance algorithms made in the context of the obstacle.” Id. at 21. In support of this construction, Patent Owner cites to various portions of the Specification. See id. at 18-21 (citing IPR2020-01475 Patent 7,228,232 B2 9 Ex. 1001, 19:27-34, 20:13-14, 20:16-49, 20:52-65, Figs. 16-20). For example, Patent Owner directs us to “[t]he method shown and described in Figures 17 and 18 results in a flight path that returns the UAV to its originally anticipated course.” Id. at 19 (citing Ex. 1001, 20:13-14, Figs. 17, 18). Patent Owner also directs us to “another obstacle avoidance algorithm taught by the ’232 patent and that can be selected from . . . a flight path that does not return the UAV to its originally anticipated course.” Id. (citing Ex. 1001, Fig. 19, 20). Patent Owner also provides testimony from Dr. Kaputa in support of this construction. Ex. 2011 ¶¶ 39-42. Petitioner takes no position on the construction of “selecting.” Pet. Reply 3. Rather, Petitioner contends “[r]esolution of the construction of ‘selecting’ is not necessary to resolve the dispute because . . . Sanders teaches ‘selecting’ under either construction.” Id. Although Petitioner provides no explicit claim construction analysis in either the Petition or the Reply, in analyzing the challenge based on Sanders, the Petition states “Sanders’ implemented DSAAV10 system includes two collision avoidance algorithms: the bang-bang maneuver and continuous maneuver.” Pet. 29. Petitioner then contends “Sanders’ algorithm selects a maneuver (bang-bang or continuous).”11 Id. at 31. Dr. Hansman likewise testifies “Sanders’ implemented DSAAV system includes two types of collision avoidance algorithms: bang-bang and continuous maneuvers” and “[t]he CAS selects one of these algorithms when an obstacle is detected.” 10 “DSAAV” refers to the Draper Small Autonomous Air Vehicle. Ex. 1005, 21. 11 Sanders describes the bang-bang maneuver and the continuous maneuver. See Ex. 1005, 68 et. seq. IPR2020-01475 Patent 7,228,232 B2 10 Ex. 1003 ¶ 101. Similarly, Petitioner applies Patent Owner’s construction of selecting in analyzing Ground 5. Pet. 70-71. Thus, both Dr. Hansman and Petitioner analyze the claims using Patent Owner’s construction that “selecting an obstacle avoidance algorithm” in claim 7 requires selection from one or more collision avoidance algorithms. Because the Petition implicitly applies the same construction proposed by Patent Owner in both of grounds 1 and 2, there is no dispute between the parties. Consequently, we evaluate the Petition using Patent Owner’s construction of “selecting.” 37 C.F.R. § 42.104 (b)(3) (the Petition must contain a statement “[h]ow the challenged claim is to be construed”); see also Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (noting that “we need only construe terms ‘that are in controversy, and only to the extent necessary to resolve the controversy’”) (citing Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)). Patent Owner also provides a table of District Court constructions of certain means plus function claim limitations recited in claims 7-12. PO Resp. 22-26. Patent Owner contends we should apply these constructions if necessary. Id. at 26. Petitioner does not dispute Patent Owner’s contention that we should apply the District Court constructions. Pet. Reply 3. Patent Owner also provides a proposed construction for a limitation of claim 9. PO Resp. 21-25. Petitioner contends Patent Owner’s proposed construction of the term in claim 9 is inconsistent with the District Court construction. Pet. Reply 4-6. We need not address the construction of any of the means plus function claim terms because it is not necessary to our Decision. See Nidec IPR2020-01475 Patent 7,228,232 B2 11 Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (noting that “we need only construe terms ‘that are in controversy, and only to the extent necessary to resolve the controversy’”) (citing Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)). D. Ground 1: Alleged Obviousness over Sanders Petitioner contends that claims 7 and 12 are unpatentable over Sanders. Pet. 18-39. In support thereof, Petitioner identifies the disclosures in Sanders alleged to describe the subject matter in the challenged claims. Id. Additionally, Petitioner offers declaration testimony from Dr. Hansman in support of the Petition. Ex. 1003 ¶¶ 113-130 Patent Owner contends that “Sanders does not teach (or suggest) the selection of an obstacle avoidance algorithm or any means therefore as required by claim 7.” PO Resp. 30. Additionally, Patent Owner offers declaration testimony from Dr. Kaputa. Ex. 2011 ¶¶ 46- 62. We begin our analysis with a brief overview of Sanders. We then address the parties’ respective contentions with respect to the challenged claims. Sanders - Ex. 1005 Sanders is titled “Real-time Collision Avoidance for Autonomous Air Vehicles” and bears a copyright date of 1998. Ex. 1005. Sanders is a master’s degree thesis describing “the design and analysis of a collision avoidance system (CAS) for autonomous air vehicles.” Id. at Abstract. The document focuses on “algorithm design, multi-AAV simulation, closed-loop analysis, and actual vehicle flight tests.” Id. Sanders discloses a “guidance, navigation, and control (GNC) system” for autonomous air vehicles IPR2020-01475 Patent 7,228,232 B2 12 (“AAV’s”). Id. at 22-23. The CAS engages in conflict detection and resolution. Id. at 55. Sanders describes two collision avoidance maneuvers. Ex. 1005, 68. The first is referred to as bang-bang maneuvers and the second is referred to as continuous maneuvers. Id. Bang-bang maneuvers “generally consist of a one-time avoid command and a one-time return command.” Ex. 1005, 68. “The return command brings the AAV back to its original trajectory after the conflict has been eliminated.” Id. Sanders describes the bang-bang maneuver as “open- loop” meaning “once it begins, it does not recalculate or readjust unless a new potential conflict has been created.” Id. Sanders discloses that the timing of the avoid command is especially important because if started “too near the time of the conflict, it may prove too late.” Id. at 69. Continuous maneuvers are closed loop meaning that they “reevaluat[e] the avoidance command at each step.” Ex. 1005, 71. Sanders discloses “[t]here are endless possibilities for continuous conflict resolution maneuvers but most seem to have one thing in common: generating optimized trajectories.” Id. Sanders describes the design of a bang-bang algorithm. Ex. 1005, 73- 78. Sanders also describes a number of limitations with the bang-bang algorithm which “sparked the design for the continuous DSAAV collision avoidance controller.” Id. at 78. IPR2020-01475 Patent 7,228,232 B2 13 Sanders describes a “[c]ontinuous [a]lgorithm” that “was designed to be an improvement over the bang-bang collision avoidance system.” Ex. 1005, 78. Sanders further describes “want[ing] to eliminate the shortcomings of the open-loop controller, especially the all-or-nothing behavior which gets it into trouble sometimes.” Id. Unlike the bang-bang algorithm, Sanders’ continuous controller “strives to send the vehicles on smooth, gradual trajectories” and “[a]ny disturbances should generally have time to correct themselves before a conflict occurs.” Id. at 79. Claim 7 Petitioner cites to various disclosures in Sanders alleged to disclose each limitation of claim 7 and also provides supporting testimony from Dr. Hansman. Pet. 18-33. Patent Owner only disputes Petitioner’s contentions with respect to the limitation “means for selecting an obstacle avoidance algorithm.” PO Resp. 28-36. We focus our analysis on this limitation. IPR2020-01475 Patent 7,228,232 B2 14 Means for Selecting an Obstacle Avoidance Algorithm As discussed above in our claim construction analysis, this limitation requires “a selection from multiple possible collision avoidance algorithms made in the context of the obstacle.” Petitioner contends Sanders’ “system includes two collision avoidance algorithms: the bang-bang maneuver and the continuous maneuver.” Pet. 29 (citing Ex. 1005, 68). Petitioner contends the bang-bang maneuver consists of a one-time avoid command and a one-time return command and “the CAS software determines whether to maneuver up or down.” Id. (citing Ex. 1005, 68, 74, 76). Petitioner further contends that the “continuous maneuver ‘strives to send the vehicles on smooth, gradual trajectories’” but like the bang-bang maneuvers “determines whether to maneuver up or down to ‘minimize[] the AAV’s deviations from their current trajectories.” Id. at 30 (citing Ex. 1005, 79, 83) (alteration in original). Based on the foregoing, Petitioner contends “Sanders’ algorithm selects a maneuver (bang-bang or continuous), parameters, and direction (e.g., up or down)” and “therefore teaches ‘selecting an obstacle avoidance algorithm.” Id. at 31 (citing Ex. 1003 ¶¶ 101-107, 127-128). Patent Owner contends “Petitioner mischaracterizes Sanders by arguing that it teaches selection of a UAV algorithm while piloting the UAV.” PO Resp. 31 (citing Pet. 28-30). According to Patent Owner, “no such selection (choice) was ever taught or suggested, because Sanders never intended that both the bang-bang algorithm and the continuous algorithm would ever be made available (e.g., for selection) concurrently on a UAV.” Id. at 31-32. Patent Owner concedes that Sanders describes both a bang- bang algorithm and a continuous algorithm. Id. at 32. In support of its IPR2020-01475 Patent 7,228,232 B2 15 contention that Sanders never intended for both to be available for selection, Patent Owner points to the disclosure in Sanders that “[t]he continuous controller was designed to be an improvement over the bang-bang collision avoidance system.” Id. at 31 (citing Ex. 1005, 78). Patent Owner further contends “the bang-bang algorithm was deemed unsuitable . . . and it was the continuous controller that Sanders thought might probably be acceptable for actual testing on a UAV.’” Id. (citing Ex. 1005, 72, 73, 78, 79). Patent Owner further contends because “Sanders intended only the continuous algorithm to ever be flown, there would never be any ‘selection’ among collision avoidance algorithms at all” as required by claim 7. Id. at 32. Patent Owner next disputes Petitioner’s contention that “the selection of a single maneuver within an obstacle avoidance algorithm meet[s] the requirements of the claims.” PO Resp. 33 (citing Pet. 31); see also Sur-reply 6 (“Choosing between ‘up’ and ‘down’ maneuvers within a running collision avoidance algorithm is not selecting an ‘obstacle avoidance algorithm.’”). Patent Owner contends that “[m]aneuvers, for example as in the case of Sander’s bang-bang algorithm, generally consist of individual one-time commands, e.g., to go up or down.” PO Resp. 34 (citing Ex. 1005, 68). Patent Owner further contends “algorithms are not one time commands” but rather “a sequence of steps for solving a logical or mathematical problem or performing a task.” Id. (citing Ex. 2005). Patent Owner contends Sanders discloses “[s]ince the [UAVs] always have two possible maneuver directions, up and down, [a] collision avoidance algorithm must set some criterion to choose between the two.” Id. at 40 (citing Ex. 1005, 76). Patent Owner further contends “more than just a maneuver, a collision avoidance algorithm requires ‘some criterion to choose’ between maneuvers and so IPR2020-01475 Patent 7,228,232 B2 16 selection of different maneuvers within a single collision avoidance algorithm cannot itself be selection from among different algorithms.” Id. (citing Ex. 2011 ¶¶ 57-60). Petitioner, in turn, contends “Sanders discloses ‘selecting’ . . . because the AAV . . . must choose between an ‘up’ and a ‘down’ algorithm when running the ‘continuous’ obstacle avoidance algorithm.” Pet. Reply 6-7 (citing Ex. 1005, 83, Fig. 4.13). In connection with Sanders’ Fig. 4.13, Petitioner contends “[t]he continuous algorithm determines whether to maneuver up . . . or down . . . to ‘minimize[] the AAV’s deviations from their current trajectories.”” Id. at 7 (citing Ex. 1005, 83, Fig. 4.13). Based on Dr. Hansman’s deposition testimony, Petitioner contends “that ‘the algorithm which guides the up maneuver would or could be different from the algorithm that provides the [down] route.’” Id. (citing Ex. 2010, 43:1- 13). Petitioner further contends “the ‘up’ and ‘down’ algorithms are not simply ‘one time commands.’” Id. Petitioner again points to Dr. Hansman’s deposition testimony that “the overall structure includes the - the logic for selecting the direction, and then there’s the sub-algorithm that then determine the trajectory - you know, how fast to pull up, how long to go, et cetera.” Id. at 7-8 (citing Ex. 2010, 43:21-44:9). Petitioner points to the obstacle avoidance algorithms in the ’232 patent cited by Patent Owner (PO Resp. 18-21) and contends Patent Owner “fails to explain, however, the distinction between what is disclosed in the ’232 patent and what is disclosed by Sanders.” Id. at 8-9. For the following reasons, we find Petitioner has not established that Sanders discloses “means for selecting an obstacle avoidance algorithm” as required by claim 7. IPR2020-01475 Patent 7,228,232 B2 17 The Petition specifically states that Sanders system includes “two collision avoidance algorithms: the bang-bang maneuver and the continuous maneuver” and “selects a maneuver (bang-bang or continuous).” Pet. 29, 31. Dr. Hansman, likewise, testifies Sanders “selects one of these algorithms when an obstacle is identified.” Ex. 1003 ¶ 101. Petitioner, however, does not dispute Patent Owner’s contention that Sanders only intended the continuous algorithm to be flown. See generally Pet. Reply. We now turn to the disclosure in Sanders concerning whether Sanders at least suggests that both collision avoidance algorithms would be available for selection during flight. Sanders discloses it anticipated running actual flight tests on vehicles with its CAS “in the coming year.” Ex. 1005, 136. The fact that Sanders had not yet engaged in any actual flight testing is not dispositive of whether Sanders teaches or suggests that both the bang-bang algorithm and the continuous algorithm would be available for selection during flight. After discussing the bang-bang algorithm, Sanders states that limitations of the bang-bang maneuver “sparked the design for the continuous DSAAV collision avoidance controller, described next.” Id. at 78. Sanders begins the description of the continuous controller by stating that “we wanted to eliminate the shortcomings of the open-loop controller, especially the all-or- nothing behavior which gets it into trouble.” Id. Sanders also states that “the continuous controller . . . will probably be used in the flight tests.” Id. at 73. Petitioner does not point to any disclosure in Sanders that it contemplated using both algorithms in upcoming flight tests nor has our independent review resulted in any disclosure from which it could be inferred that Sanders intended to include both algorithms during flight tests. IPR2020-01475 Patent 7,228,232 B2 18 In light of the foregoing, we find Petitioner has not shown that Sanders teaches or suggests that two collision avoidance algorithms would be available for selection during flight. In the absence of such a teaching or suggestion, to the extent Petitioner contends Sanders teaches a selection between the bang-bang algorithm and the continuous algorithm, Petitioner fails to show that Sanders discloses “means for selecting a collision avoidance algorithm” as required by claim 7. In the Reply, Petitioner focuses on the disclosure in Sanders that purportedly shows Sanders “must choose between an ‘up’ and a ‘down’ algorithm when running the ‘continuous’ obstacle avoidance algorithm.” Pet. Reply 6-7. The Petition states Sanders’ continuous algorithm determines “whether to maneuver up or down.” Pet. 30 (citing Ex. 1005, 76, 83). The Petition also references the up and down maneuvers as part of its contention that Sanders discloses selecting an obstacle avoidance algorithm. See id. at 31. Dr. Hansman offers testimony to the same effect. Ex. 1003 ¶ 107. In order to support the contention in the Reply, Petitioner relies on Dr. Hansman’s deposition testimony “that ‘the algorithm which guides the up maneuver would or could be different from the algorithm that provides the [down] route.” Pet. Reply 7 (citing Ex. 2010, 43:1-13). Dr. Hansman did not provide any testimony in his declaration that the up and down maneuvers in Sanders constitute separate collision avoidance algorithms. See Ex. 1003 ¶¶ 101-107. Further neither Petitioner nor Dr. Hansman cites to any disclosure in Sanders to support this theory. Dr. Kaputa testifies “a [skilled artisan] would not use the words maneuver and algorithm interchangeably. Indeed, not even Sanders does so IPR2020-01475 Patent 7,228,232 B2 19 and instead makes clear that a collision avoidance algorithm includes maneuvers.” Ex. 2011 ¶ 57 (citing Ex. 1005, 141 et seq.). Dr. Kaputa further testifies “a maneuver can be executed via an algorithm, however, in the case of Sanders there is a clear delineation between maneuvers and collision avoidance algorithms.” Id. Dr. Kaputa further testifies: Sanders further states “Many collision avoidance algorithms in the literature seem to choose a maneuver direction with little thought of justification. As we will see, however, this decision can depend on many factors. Collision avoidance maneuvers usually fall into one of four categories: horizontal maneuvers, vertical maneuvers, speed changes, and combination maneuvers. EX. 1005 at 65 (emphasis in original). Further, “Most algorithms in the collision avoidance literature seem to use horizontal resolution maneuvers.” Id. at 65-66. Thus, it is clear that for a [skilled artisan], a collision avoidance maneuver is not a collision avoidance algorithm, rather it is an instruction contained within a higher level collision avoidance algorithm that may be performed under the right circumstances. Id. ¶ 60. Petitioner submits no evidence to rebut this testimony by Dr. Kaputa. Further, this testimony is consistent with Dr. Hansman’s declaration testimony. See Ex. 1003 ¶ 106 (testifying the “the continuous algorithm determines when to maneuver up . . . or down”). In addition, Sanders discloses its continuous algorithm contains a step where it “Compute[s] collision avoidance maneuver.” Ex. 1005, 149. We therefore credit Dr. Kaputa’s testimony because it is based on corroborating evidence from Sanders over Dr. Hansman’s unsupported deposition testimony. To the extent that Petitioner contends that these maneuvers correspond to collision avoidance algorithms in the ’232 patent (Pet. Reply 8-9), Petitioner’s contention is not persuasive because it is attorney argument not supported by evidence. Consequently, Petitioner fails to establish with evidence that IPR2020-01475 Patent 7,228,232 B2 20 Sanders’ continuous algorithm choosing between an up or down maneuver satisfies the limitation “selecting a collision avoidance algorithm.” For all the foregoing reasons, Petitioner has not shown that Sanders discloses or teaches the limitation “means for selecting an obstacle avoidance algorithm.” Consequently, we determine Petitioner fails to establish by a preponderance of the evidence that claim 7 of the ’232 patent is unpatentable over Sanders. Claim 12 depends from claim 7. Ex. 1001, 22:38-43. We reviewed Petitioner’s contentions and evidence cited for claims 12. Pet. 33-39. The contentions and evidence do not cure the deficiencies in claim 7 discussed above. Consequently, we determine that Petitioner fails to establish by a preponderance of the evidence that claim 12 of the ’232 patent is unpatentable over Sanders. E. Grounds 2, 3, and 4 Claims 8-11 depend from claim 7. Ex. 1001, 22:4-37. Petitioner details the disclosure in Sanders, Pilley, Sutter, and Pappas that it contends corresponds to the limitations in claims 8-11 as well as reasons for combining Sanders with each of Pilley, Sutter, and Pappas. Pet. 39-57. We reviewed Petitioner’s contentions as well as the evidence cited for these dependent claims and find it does not cure the deficiencies in claim 7 discussed above. Consequently, we determine that Petitioner fails to establish that claims 8-11 of the ’232 patent are unpatentable over Sanders in combination with the other relied-upon prior art. IPR2020-01475 Patent 7,228,232 B2 21 F. Ground 5: Obviousness over Duggan and Pilley Petitioner contends that claims 7-9 and 12 are unpatentable over Duggan and Pilley. Pet. 57-82. In support thereof, Petitioner identifies the disclosures in Duggan and Pilley alleged to describe the subject matter in the challenged claims. Id. Additionally, Petitioner offers declaration testimony from Dr. Hansman. Ex. 1003 ¶¶ 224-326. Patent Owner contends a skilled artisan would not have been motivated to combine Duggan and Pilley as proposed. PO Resp. 40-47. We begin our analysis with a brief overview of Duggan and Pilley. We then address the parties’ respective contentions with respect to the challenged claims. Duggan - Ex. 1012 Duggan discloses “a variable autonomy control system that enables a human to manage and operate a vehicle through interaction with a human- system interface.” Ex. 1012, 1:14-17. The embodiments disclosed in Duggan are described “in the context of an unmanned aerial vehicle (UAV).” Id. at 4:53-54. Duggan’s variable autonomy control system (VACS) provides for control of a UAV “from fully autonomous control to simplified manual flight control modes for enhanced real-time control.” Id. at 5:19-25. Duggan’s VACS comprises a ground station connected by a data link to a UAV. Id. at Fig. 6. Executive component 652 located on airborne system interface 604 “process[es] commands and information received from command control component 606” and manages UAV components such as navigation 662, guidance 658, and autopilot 654. Id. at 15:64-66. Among the functions performed by Duggan’s system are autonomous ground collision avoidance and autonomous see-and-avoid air IPR2020-01475 Patent 7,228,232 B2 22 collision avoidance. Id. at 7:48-53, Fig. 1. Ground collision avoidance system 120 “provides an automated mechanism enabling the control system to avoid terrain” and air collision avoidance system 122 autonomously avoids collisions with other aircraft. Id. at 11:8-11, 43:12-17. Pilley -- Ex. 1010 Pilley is directed to a system for controlling surface and airborne traffic using GNSS12-based or GPS-based data. Ex. 1010, 4:11-14. Pilley’s system controls and manages the flow of traffic approaching and departing an airport as well as the flow of surface vehicles and taxiing aircraft. Id. at 4:24-27. Pilley defines a dynamic zone around an aircraft that moves with the aircraft. Id. at 11:11-17, Fig. 4. Pilley performs a rough check for vehicles in the immediate vicinity and uses a detection algorithm that “project[s] the position of a vehicle ahead by an increment of Time(t) using the received vehicle velocity.” Id. at 94:1-4, 297:32-34, Fig. 10. If the distance between vehicles at projected distances is less than a safe distance, Pilley generates an alert of a potential collision. Id. at 94:8-21. Claim 7 a. Preamble Petitioner contends that “Duggan’s VACS segregates UAV control into two fundamental categories: flight control associated ‘with the aviation of the aircraft’ and flight management associated ‘with the mission plan (navigation tasking) for the aircraft.” Pet. 65 (citing Ex. 1012, 6:1-6). Based on this, Petitioner contends that “the combination of Duggan and 12 “GNSS” refers to the Global Navigation Satellite System. Ex. 1010, 3:38-39. IPR2020-01475 Patent 7,228,232 B2 23 Pilley discloses a ‘[a] system for navigating a UAV.’” Id. (citing Ex. 1003, ¶¶ 270-271) (alteration in original). Patent Owner does not squarely address Petitioner’s contentions or whether the preamble is limiting. See PO Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find that Duggan discloses the subject matter of the preamble. We do not reach the question of whether the preamble is limiting because it is not necessary to our determination. b. Means for Piloting the UAV Under Control of a Navigation Computer, in Accordance with a Navigation Algorithm Petitioner contends that Duggan discloses “a baseline set of guidance laws” such as “waypoint guidance law 1134; and loiter guidance mode 1142.” Pet. 68 (citing Ex. 1012, 21:33-35, 51-58). Petitioner further contends that “Duggan’s set of guidance laws collectively is the recited ‘navigation algorithm.’” Id. (citing Ex. 1003 ¶ 243). Petitioner further contends that “Duggan’s guidance component ‘is responsible for generating autopilot commands that, when executed, achieve a particular guidance objective.’” Id. at 69 (citing Ex. 1012, 21:22-24). Petitioner further contends that “Duggan’s autopilot 654 . . . ‘receives flight control commands, processes the commands . . . and passes the commands to actuator control component 668 and engine control component 666 as necessary for flight control execution.’” Id. (citing Ex. 1012, 16:9-14). According to Petitioner, “Duggan’s autopilot therefore pilots the UAV under control of the guidance component (‘navigation computer’) in accordance with the guidance laws (‘navigation algorithm’).” Id. (citing Ex. 1003 ¶ 244). Petitioner further contends that the “UAV processor executing Duggan’s autopilot performs the ‘piloting’ function ‘under control of a navigation computer, in accordance with a navigation algorithm’ and is IPR2020-01475 Patent 7,228,232 B2 24 therefore the recited ‘means for piloting.’” Id. (citing Ex. 1003 ¶¶ 240-244, 272-274). Patent Owner does not squarely address Petitioner’s contentions. See P) Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find Duggan discloses this limitation. c. Means for Reading from a GPS receiver a sequence of GPS Data Petitioner contends that Duggan discloses that its flight control system can be configured to support a GPS receiver. Pet. 69 (citing Ex. 1012, 58:1- 4, 17:39-44). Petitioner further contends that Duggan’s navigation component 662 “facilitates the collection of data pertaining to aircraft location.” Id. at 70 (citing Ex. 1012, 16:7-9, Fig. 6). Petitioner further contends that the information obtained by navigation component 662 is “used by the guidance functionality to aid in decision-making such as, but certainly not limited to, path-regulation.” Id. (citing Ex. 1012, 21:12-20). Petitioner further contends that “GPS data is continuously generated and ‘read’ by the navigation and guidance components to perform collision avoidance and flight management functions.” Id. (citing Ex. 1003 ¶ 248; Ex. 1012, 17:51-57, Fig. 8). Based on the foregoing, Petitioner contends that “Duggan’s systems ‘read[s] . . . a sequence of GPS data.’” Id. (citing Ex. 1003 ¶¶ 248-249) (alteration in original). Petitioner further contends that “Pilley also uses GPS data for ‘position projections, coordinate conversions, zone detection, collision prediction, [and] runway incursion detection’ among other functions.” Pet. 71 (citing Ex. 1010, 7:33-35) (alteration in original). Petitioner further contends that “Pilley teaches that ‘[c]ollision processing is performed each second, upon receipt of the FEV’s [fully equipped vehicle] GPS position IPR2020-01475 Patent 7,228,232 B2 25 and velocity data.’” Id. (citing Ex. 1010, 328:25-32) (alterations in original). Based on the foregoing, Petitioner contends that “Pilley’s system also ‘read[s] . . . a sequence of GPS data.’” Id. (citing Ex. 1003 ¶ 249) (alterations in orginal). Petitioner further contends that the “UAV processor executing the navigation and guidance performs the ‘reading’ function and is therefore the recited ‘means for reading.’” Id. (citing Ex. 1003 ¶¶ 245- 249, 275-277). Patent Owner does not squarely address Petitioner’s contentions. See PO Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find Duggan and Pilley disclose this limitation. d. Means for Anticipating a Future Position of the UAV in Dependence Upon the Sequence of GPS Data Petitioner contends that “Duggan’s ‘ground collision avoidance system (GCAS) provides an automated mechanism enabling the control system to avoid terrain.” Pet. 72 (citing Ex. 1012, 43:12-17). Petitioner further contends that Duggan reads a vehicle’s position from GPS data and its’ “GCAS algorithm predicts a collision by generating a scan pattern that ‘starts at the vehicle position.” Id. (citing Ex. 1012, 43:54-55). Petitioner further contends that Duggan generates a horizontal uncertainty box and the GCAS algorithm calculates the maximum terrain altitude in each uncertainty box. Id. (citing Ex. 1012, 43:55-59, Fig. 23). Petitioner further contends that the “GCAS algorithm next “assumes a constant initial velocity and projects the trajectory along the velocity line to estimate a time to fly up.” Id. (citing Ex. 1012, 44:23-25). Petitioner further contends that “[i]n an embodiment, Pilley’s detection process is performed onboard the aircraft.” Id. at 60 (citing Ex. 1010, 93:47-52) (emphasis added). Petitioner further contends that “[t]he modified air collision avoidance component of the IPR2020-01475 Patent 7,228,232 B2 26 combined system integrates Pilley’s GPS-based collision detection” and that “Pilley teaches that collision detection performed on the vehicle ‘us[es] the current GPS data and the information stored in the vehicle database.” Id. at 73 (citing Ex. 1010, 297:19-21) (alteration in original). Petitioner further contends that Pilley’s algorithm “performs . . . collision checking by projecting the aircraft’s position ahead in one second intervals.” Id. at 73- 74 (citing Ex. 1010, 297:36-38). Petitioner further contends that the “UAV processor executing the GCAS and modified air collision avoidance algorithms performs the function of ‘anticipating a future position of the UAV in dependence upon the sequence of GPS data’ and is therefore the recited ‘means for anticipating.’” Id. at 74 (citing Ex. 1003 ¶¶ 250-258, 279). Patent Owner contends that Pilley’s “air collision avoidance is a ground control directed function” and “Pilley does not even report an embodiment in which the collision detection determination is reported on board an aircraft.” PO Resp. 43. In support of this contention, Patent Owner directs us to “Pilley’s Vehicle Functional Matrix” wherein, according to Patent Owner, “dynamic collision processing is not performed on aircraft. Instead, it is performed at the ground-based AC&M13 system, and on fully equipped ground vehicles.” Id. (citing Ex. 1010, 325:25-53, 327:6-12, 328:20-35). Petitioner, in turn, contends Pilley discloses that collision prediction would be on board the aircraft/vehicles. Pet. Reply 16 (citing Ex. 1010, 3:34-44). Petitioner further contends “Pilley’s Vehicle Functional Matrix 13 “AC&M” refers to an Airport Control and Management System. Ex. 1010, 4:48. IPR2020-01475 Patent 7,228,232 B2 27 relates to ‘prototype demonstrations’ of ‘[t]hree vehicles, equipped with varying configurations of hardware and software.” Id. at 17 (citing Ex. 1010, 325:19-22) (alteration in original). Petitioner further contends the “prototypes are only a single example of Pilley’s disclosure.” Id. For the following reasons, we are persuaded by Petitioner’s contentions. First, we note claim 7 does not require performing collision detection on the aircraft. Second, Pilley’s Vehicle Functional Matrix relates to “prototype demonstrations” of “[t]hree vehicles, equipped with varying configurations of hardware and software.” Ex. 1010, 325:19-22. The matrix indicates that the aircraft used in the prototype demonstrations did not “Perform[] dynamic collision processing” but it also indicates that the aircraft did “Perform[] zone incursion processing.” Id. at 325:39-42. The description of the matrix does not suggest that Pilley’s system is limited to the specific details of the prototype vehicles. See id. at 325:15-327:33. Pilley discloses that it anticipates its system is “capable of performing the navigation, surveillance, collision prediction, and zone/runway incursion . . . on the aircraft.” Ex. 1010, 3:41-42. Pilley further discloses dynamic zones around aircraft that move with the aircraft. Id. at 11:11-13, Fig. 4. The zones shown in Pilley’s Figure 10 are used in the collision detection system described in Pilley. See id. at 93:41-94:45, 297:16-63, Fig. 10. Although the prototype demonstrations reported in the Vehicle Functional Matrix indicate that collision processing was not performed on board the aircraft, the disclosure just discussed suggests to a skilled artisan that Pilley contemplated equipping the aircraft with collision detection. IPR2020-01475 Patent 7,228,232 B2 28 Although Petitioner’s evidence supports a finding that Duggan and Pilley disclose this limitation, as discussed below in connection with Motivation to Combine, we find Petitioner has not established that a skilled artisan would have combined Duggan and Pilley as proposed in the Petition. e. Means for Identifying an Obstacle in Dependence Upon the Future Position Petitioner contends that “Duggan’s GCAS algorithm ‘provides the operator with some form of situational awareness information about an upcoming collision (e.g., a warning signal).’” Pet. 74 (citing Ex. 1012, 43:17-20). Petitioner further contends that “GCAS ‘monitors vehicle altitude and ground altitude attempts to predict at what point the vehicle will intersect the terrain.” Id. (citing Ex. 1012, 44:1-3). Petitioner further contends that the “air collision avoidance component” in Petitioner’s proposed combination “determines whether the UAV is within a safe clearance distance of other vehicles at projected future positions” and “[w]hen the separation distance between the UAV and another vehicle is less than the safe distance clearance, a potential collision alert is generated.” Id. at 75 (citing Ex. 1010, 94:8-14, 297:42-45). According to Petitioner, “the algorithm identifies whether the vehicle is an ‘obstacle’ at a future position of the UAV. Id. (citing Ex. 1003 ¶ 261). Petitioner further contends that the “UAV processor executing the GCAS and modified air collision avoidance algorithms performs the function of ‘identifying an obstacle in dependence upon the future position’ and is therefore the recited ‘means for identifying.’” Id. (citing Ex. 1003 ¶¶ 259-261, 281). Patent Owner does not squarely address Petitioner’s contentions. See PO Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find Duggan and Pilley disclose this limitation. IPR2020-01475 Patent 7,228,232 B2 29 f. Means for Selecting an Obstacle Avoidance Algorithm Petitioner contends that if Duggan detects terrain as an obstacle, “the system selects GCAS which provides an automated mechanism to avoid the terrain.” Pet. 75 (citing Ex. 1012, 43:13-15). Petitioner further contends that “[i]f the obstacle is another vehicle, the system selects air collision avoidance” which “override[s] the waypoint path and direct[s] the aircraft off of the collision course.” Id. at 76 (citing Ex. 1003 ¶ 264; Ex. 1012, 11:6-11). Petitioner further contends that the “UAV processor executing the collision avoidance software performs the function of ‘selecting an obstacle avoidance algorithm’ and is therefore the ‘means for selecting.’” Id. (citing Ex. 1003 ¶¶ 263-265, 283). Patent Owner does not squarely address Petitioner’s contentions. See PO Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find Duggan and Pilley disclose this limitation. g. Means for Piloting the UAV in Accordance with the Selected Obstacle Avoidance Algorithm Petitioner contends that both of Duggan’s GCAS and air collision avoidance components autonomously execute obstacle avoidance algorithms. Pet. 76. Petitioner contends that “[f]or ground collision avoidance, the system causes the autopilot to execute a vertical fly-up maneuver to ‘automatically avoid the terrain.’” Id. (citing Ex. 1012, 43:20- 23). Petitioner further contends that for air collisions, “Duggan’s collision avoidance commands ‘will first correspond to the requested waypoint path, then there will be a transition to a non-collision path, and then there will be a graceful transition back to the waypoint path.’” Id. (citing Ex. 1012, 11:11- 15). Petitioner further contends that the “UAV processor executing the GCAS and modified air collision avoidance algorithms performs the IPR2020-01475 Patent 7,228,232 B2 30 function of ‘piloting the UAV in accordance with the selected obstacle avoidance algorithm’ and is therefore the recited ‘means for piloting.’” Id. at 77 (citing Ex. 1003 ¶¶ 266-268, 285). Patent Owner does not squarely address Petitioner’s contentions. See PO Resp. 40-47. Based on our review of the evidence cited by Petitioner, we find Duggan and Pilley disclose this limitation. h. Motivation to Combine Petitioner contends a skilled artisan “would have been motivated to modify Duggan’s air collision avoidance component to integrate Pilley’s air collision detection processes.” Pet. 62 (citing Ex. 1003 ¶¶ 235-238). Petitioner provides the following annotated version of Duggan’s Figure 1: Id. at 63. Figure 1 of Duggan is a block diagram illustrating Duggan’s variable autonomy control system. Ex. 1012, 3:40-41. Petitioner highlights in red block 122 of Figure 1 illustrating what Petitioner terms “the modified air collision avoidance component module.” Pet. 63. Petitioner’s proposed IPR2020-01475 Patent 7,228,232 B2 31 modification of Duggan is based on plugging the air collision avoidance module of Pilley into block 122 in accordance with “Duggan’s plug-n-play architecture.” Id. at 64 (citing Ex. 1012, 7:12-15, 8:57-61). Petitioner provides several reasons why a skilled artisan would have purportedly integrated Pilley’s air collision avoidance processes into Duggan. First, Petitioner contends “Duggan does not provide any details regarding the mechanism used to detect potential obstacles or collisions other than a brief mention of an optical tracker” and a skilled artisan would therefore “look for references disclosing details of air collision detection techniques and would have been led to Pilley.” Pet. 63 (citing Ex. 1003 ¶ 236). Second, Petitioner contends a skilled artisan “would have been motivated to integrate GPS-based air collision detection into Duggan to provide an enhanced collision avoidance system that avoids problems of optical systems due to environmental conditions (e.g., clouds, fog, etc.).” Id. at 63-64 (citing Ex. 1003 ¶ 237). Petitioner further contends Pilley provides motivation for the combination by touting “the application of new technologies to the management of our airports [to] provide improved efficiency, enhanced safety.” Id. at 64 (citing Ex. 1010, 3:45-50). Third, Petitioner contends Duggan “stresses that collision avoidance ‘and other applicable functions are commercially available modules for integration’” and “can be integrated in a ‘plug-n-play’ format.” Id. (citing Ex. 1012, 7:12-15, 7:42-44, 8:57-61). Petitioner further contends it would have “been obvious to a [skilled artisan] to integrate modified or additional collision avoidance capabilities to enhance the operation and safety of Duggan’s system.” Id. (citing Ex. 1003 ¶ 238). IPR2020-01475 Patent 7,228,232 B2 32 Patent Owner contends that a skilled artisan would not have modified Duggan’s control system as proposed because of the differences between Duggan’s UAV collision avoidance process and Pilley’s airport collision avoidance process. PO Resp. 40-47. Patent Owner contends Duggan discloses “a variable autonomy control system that enables a human to manage and operate a vehicle [e.g., a UAV] through interaction with a human-system interface.” PO Resp. 40 (citing Ex. 1012, 1:14-17, 4:53-54). Patent Owner further contends Duggan’s “variable autonomy control system (VACS) includes a ground station connected by a data link to a UAV.” Id. (citing Ex. 1012, Fig. 6). Patent Owner further contends “Duggan distinguishes between UAV flight control and UAV flight management.” Id. at 47 (citing Ex. 1012, 5:67-6:6). Patent Owner further contends that in Duggan, “mission control is a UAV autonomous function” and air collision avoidance is part of Duggan’s mission control. Id. at 41 (citing Ex. 1012, 7:48-53, 9:34-41; Ex. 1020, 55:21-56:21; Ex. 2010, 56:22-57:22). According to Patent Owner, because air collision avoidance is a UAV autonomous function, “it is a decision that the UAV needs to make for itself based on what it perceives from its environment and what it has been programmed to view as a condition requiring some action.” Id. (citing Ex. 2010, 57:16-22). Patent Owner next contends “for Pilley, air collision avoidance is not an aircraft autonomous function.” PO Resp. 42. According to Patent Owner, in Pilley, “surface (i.e., ground) vehicles and aircraft report their positions to a ground control station.” Id. (citing Ex. 1012, 291:5 et seq., Fig. 22). Patent Owner further contends “[b]oth the surface vehicles and aircraft send Automatic Dependent Surveillance (ADS) messages to AC&M IPR2020-01475 Patent 7,228,232 B2 33 processor 115, which in turn composes air traffic control (‘ATC’) messages that are forwarded to the vehicles and aircraft.” Id. Patent Owner further contends Pilley’s “AC&M system uses information obtained from the ADS messages to construct a database containing the positions and movements of surface vehicles and aircraft” and the AC&M system uses this information to predict collisions. Id. (citing Ex. 1012, 9:32-39, 9:44-45). Patent Owner further contends “[t]he ATC messages are used to provide the vehicles and aircraft sufficient information to avoid such collisions.” Id. (citing Ex. 1012, 293:10-20, 293:42-45; Ex. 2010, 77:16-79:3). Based on this, Patent Owner asserts Pilley’s “air collision avoidance is a ground control directed function.” Id. at 49. Patent Owner next contends a skilled artisan would not have been motivated “to implement the teachings of Pilley regarding a ground control directed collision avoidance system in a system such as that described by Duggan where collision avoidance is required to be a function a UAV performs itself” because “[r]elying on a ground station for direction concerning collision avoidance changes the principle of operation . . . of Duggan because a UAV would no longer be self-reliant for collision avoidance decisions.” PO Resp. 44 (citing Ex. 2011 ¶¶ 72, 74, 76-78). Patent Owner further contends relying on inputs from other aircraft to make collision avoidance decisions “opens the UAV to spoofing and rather than avoiding collisions, may in fact lead to collisions if false information is provided.” Id. at 45 (citing Ex. 2017, 2-20). Patent Owner next contends “Sanders describes the dangers associated with adopting just any collision avoidance system, especially one which requires cooperation of other aircraft: incorrect assumptions may lead to IPR2020-01475 Patent 7,228,232 B2 34 collisions.” PO Resp. 45 (citing Ex. 1005, 76). Patent Owner further contends that Petitioner proposed such a cooperative system where “all vehicles rely on the others sending information to a central controller and executing commands issued by a central controller” which Patent asserts “eliminates an autonomous see-and-avoid system that Duggan advocates for its collision avoidance system and which others deemed necessary.” Id. (citing Ex. 1012, 7:48-53; Ex. 2017, 2-3). Petitioner, in turn, contends Patent Owner’s arguments against the Duggan Pilley combination fail because “Pilley discloses collision prediction onboard an aircraft.” Pet. Reply 16 (citing Pet. 60-62). Petitioner next contends Patent Owner has no basis “to support its conclusory allegation that Duggan’s principle of operation is ‘self- relian[ce]’” because, according to Petitioner, “Duggan’s intent was to create a control system that could support ‘multiple levels of autonomous operation.’” Id. at 18 (citing Ex. 1012, 1:56-67) (alteration in original). Petitioner next contends “Sanders does not teach away from the use of any and all cooperative systems. Sanders identified limitations with the specific bang-bang algorithm.” Id. at 19 (citing Ex. 1005, 78). According to Petitioner, Sanders “did not--and could not have--taught away from a system that was not actually disclosed until years later.” Id. at 19-20 (citing Ex. 1005, 1; Ex. 1012, code (22)). For the following reasons, we determine that Petitioner has not established that a skilled artisan “would have been motivated to modify Duggan’s air collision avoidance component to integrate Pilley’s air collision detection processes.” Pet. 62. IPR2020-01475 Patent 7,228,232 B2 35 First, Petitioner and Dr. Hansman rely heavily on Duggan’s disclosure that its VACS architecture is “plug-n-play.” Pet. 58; Ex. 1003 ¶ 238; see also Tr. 9:12-20 (arguing “how simple the combination would be”, “Duggan has designed such that additional components can be modularly plugged into the control architecture”, “Duggan talking about how things are just plug and play”). This argument appears to assume that a skilled artisan is an automaton and any modular collision avoidance system that could be plugged in would suffice for obviousness. However, the test is not whether a skilled artisan could plug Pilley’s air collision detection process into Duggan but rather whether a skilled artisan would have done so. In that regard, Dr. Kaputa testifies to the “difference between creating a modular function that ‘plugs-n-plays’ with an intelligent synthesizer (as discussed by Duggan) and a component such as a GPS-based collision avoidance system that relies on zone databases constructed from time varying position and state information provided by other aircraft in the region.” Ex. 2011 ¶ 80. Among the reasons cited by Dr. Kaputa to support his opinion of a difference is a skilled artisan would “have to ensure that all aircraft flying within the vicinity of the UAV also adhere, subscribe, and transmit data to the new GPS-based collision avoidance system in order for it to be effective.” Id. (citing Ex. 1010, 93:41-94:60); see also Ex. 1005, 76 (“It is especially important for each vehicle’s CAS to know whether the other vehicle are cooperating with them in collision avoidance.”). Second, Patent Owner provides evidence concerning see-and-avoid collision avoidance requirements for a UAV as required in Duggan (Ex. 1012, 7:47-50). Patent Owner directs us to Exhibit 2017 which is titled “Issues Concerning Integration of Unmanned Aerial Vehicles in Civil IPR2020-01475 Patent 7,228,232 B2 36 Airspace.” Ex. 2017, cover (“DeGarmo”). In particular, DeGarmo discusses see-and-avoid collision avoidance for UAVs and points to a distinction between see-and-avoid collision avoidance requirements in manned vs. unmanned vehicles. Id. at 2-3. DeGarmo discloses: To avoid collisions, UAV’s must have a “see and avoid” capability . . . that allows them to detect and safely steer clear of aircraft or other obstructions. The pilot’s responsibilities in see-and-avoid activities in manned aircraft are spelled out in FAA Advisory circular 90-48C, Pilot’s Role in Collision Avoidance. For UAVs, FAA directive 7610.4J, entitled Special Military Operations, states that the UAV (referred to as “ROAs” in the directive) operations require “the proponent to provide the ROA with a method that provides an equivalent level of safety, comparable to see-and-avoid requirements for manned aircraft.” This in essence, ties it to AC90-48C. Further, FAR Part 91.113, Right of Way Rules, states that: “regardless of whether an operation is conducted under instrument flight rules or visual flight rules, vigilance shall be maintained by each person operating an aircraft as to see and avoid other aircraft.” To satisfy the requirements, all UAVs must therefore be able to reliably avoid collisions with all aircraft--cooperative and non-cooperative--at all times. Id. DeGarmo notes that UAVs are essentially tethered to ground based links that “are used for vehicle control, monitoring, and air traffic communications and are, to varying degrees vulnerable to jamming, spoofing, and interference or attempts to usurp control.” Ex. 1017, 2-20. Sanders, like DeGarmo, refers to the need for cooperative communication between UAVs to avoid collisions. Ex. 1005, 78. Petitioner fails to address in the Petitioner Reply the distinction in the FAA see-and-avoid requirements for manned and unmanned vehicles discussed in DeGarmo. See Pet. Reply 16-20. With respect to Sanders, Petitioner argues that “Sanders does not teach away from the use of any and IPR2020-01475 Patent 7,228,232 B2 37 all cooperative systems.” Id. at 19. Patent Owner, however, does not argue that Sanders teaches away from using cooperative communications. Rather, based on our review of Sanders, we find that Sanders corroborates, in part, the reasons why DeGarmo’s teachings undercut Petitioner’s reasons for plugging Pilley’s collision avoidance processing into Duggan’s architecture. Ex. 1005, 76. Petitioner focuses much of its Reply on the fact that Pilley teaches it contemplates putting its collision avoidance system on an aircraft. Pet. Reply 16-19. Petitioner, however, does not direct us to any disclosure in Pilley that it intended its collision avoidance system to be used on a UAV.14 See generally Pet.; Pet. Reply; Tr. 11:7-11. Further, Petitioner fails to provide persuasive evidence that Pilley’s manned vehicle collision avoidance system wherein, as disclosed in DeGarmo, the pilot plays a role in see-and-avoid collision avoidance, would provide an equivalent level of safety in an unmanned UAV as required by the FAA requirements referenced in DeGarmo (Ex. 2017, 2-3). Given the distinction between see- and-avoid collision requirements in UAVs as opposed to manned aircraft as disclosed in DeGarmo, Petitioner does not persuade us that a skilled artisan would have simply plugged Pilley’s collision avoidance processing into Duggan’s architecture. For all the foregoing reasons, we determine Petitioner fails to establish that claim 7 of the ’232 patent is unpatentable over Duggan and Pilley. 14 Petitioner does not assert that it would have been obvious to modify Pilley’s collision avoidance processing for use in a UAV, IPR2020-01475 Patent 7,228,232 B2 38 Claims 8, 9, and 12 Claims 8, 9, and 12 depend from claim 7. Petitioner details the disclosure in Duggan and Pilley that it contends corresponds to the limitations in claims 8, 9, and 12. Pet. 77-82. We reviewed Petitioner’s contentions as well as the evidence cited for these dependent claims and find it does not cure the deficiencies in claim 7 discussed above. Consequently, we determine that Petitioner fails to establish that claims 8, 9, and 12 of the ’232 patent are unpatentable over Duggan and Pilley. G. Ground 6: Obviousness over Duggan, Pilley, and Sainthuile Claims 10 and 11 depend from claim 7. Petitioner details the disclosure in Duggan, Pilley, and Sainthuile that it contends corresponds to the limitations in claims 10 and 11 as well as reasons for combining Duggan, Pilley, and Sainthuile. Pet. 82-91. We reviewed Petitioner’s contentions as well as the evidence cited for these dependent claims and find it does not cure the deficiencies in claim 7 discussed above. Consequently, we determine that Petitioner fails to establish that claims 10 and 11 of the ’232 patent are unpatentable over Duggan and Pilley. III. CONCLUSION Based on the record before us, we determine the following: IPR2020-01475 Patent 7,228,232 B2 39 IV. ORDER In consideration of the foregoing, it is hereby: ORDERED that, based on the preponderance of the evidence, claims 7-12 of the ’232 patent have not been shown to be unpatentable; and FURTHER ORDERED that, because this is a Final Written Decision, parties to this proceeding seeking judicial review of this Decision must comply with the notice and service requirements of 37 C.F.R. § 90.2. Claims 35 U.S.C. § Reference(s)/Basis Claims Shown Unpatentable Claims Not shown Unpatentable 7, 12 103(a) Sanders 7, 12 8 103(a) Sanders, Pilley 8 9 103(a) Sanders, Suiter 9 10, 11 103(a) Sanders, Pappas 10, 11 7-9, 12 103(a) Duggan, Pilley 7-9, 12 10, 11 103(a) Duggan, Pilley, Sainthuile 10, 11 Overall Outcome 7-12 IPR2020-01475 Patent 7,228,232 B2 40 FOR PETITIONER: Joshua L. Goldberg joshua.goldberg@finnegan.com Qingyu Yin qingyu.yin@finnegan.com Kelly Lu kelly.lu@finnegan.com Benjamin Benjamin Saidman benjamin.saidman@finnegan.com Yin Cheung yin.cheung@dji.com Luke MacDonald luke.macdonald@finnegan.com FOR PATENT OWNER: Tarek Fahmi tarek.fahmi@ascendalaw.com Holly Atkinson holly.atkinson@ascendalaw.com Jonathan Tsao jonathan.tsao@ascendalaw.com