Autel Robotics USA LLCDownload PDFPatent Trials and Appeals BoardMay 13, 2020IPR2019-00250 (P.T.A.B. May. 13, 2020) Copy Citation Trials@uspto.gov Paper 29 Tel: 571-272-7822 Date: May 13, 2020 UNITED STATES PATENT AND TRADEMARK OFFICE ____________________ BEFORE THE PATENT TRIAL AND APPEAL BOARD ____________________ SZ DJI TECHNOLOGY CO., LTD. Petitioner, v. AUTEL ROBOTICS USA LLC, Patent Owner. IPR2019-00250 Patent 7,979,174 B2 Before ERICA A. FRANKLIN, JENNIFER MEYER CHAGNON, and AVELYN M. ROSS, Administrative Patent Judges. ROSS, Administrative Patent Judge. JUDGMENT Final Written Decision Determining All Challenged Claims Unpatentable 35 U.S.C. § 318(a) IPR2019-00250 Patent 7,979,174 B2 2 I. INTRODUCTION SZ DJI Technology, Co. Ltd. (“Petitioner”) filed a Petition requesting inter partes review of claims 9–17 of U.S. Patent No. 7,979,174 B2 (Ex. 1001, “the ’174 patent”). Paper 2 (“Pet.”). Autel Robotics USA LLC (“Patent Owner”) did not file a Preliminary Response to the Petition. See 37 C.F.R. § 42.107(b) (2018) (setting the due date for filing a preliminary response). Upon consideration of the Petition and the evidence of record, we determined that Petitioner had demonstrated a reasonable likelihood that it would prevail with respect to at least one claim of the ’174 patent. Paper 6 (“Dec.”). Thus, we instituted review of all challenged claims on all grounds. Id. Following institution of trial, Patent Owner filed a Patent Owner Response (Paper 12, “PO Resp.”), Petitioner filed a Reply (Paper 18, “Reply”), and Patent Owner filed a Sur-reply (Paper 20, “Sur-reply”). In support of their respective positions, Petitioner relies on the testimony of Dr. Alfred Ducharme (Ex. 1003), and Patent Owner relies on the testimony of Dr. Charles F. Reinholtz (Ex. 2001). An oral hearing was held on February 19, 2020, and a transcript of the hearing is included in the record (Paper 28, “Tr.”). We have jurisdiction under 35 U.S.C. § 6. This Final Written Decision is issued pursuant to 35 U.S.C. § 318(a) and 37 C.F.R. § 42.73. For the reasons discussed below, we determine that Petitioner has shown by a preponderance of the evidence that claims 9–17 of the ’174 patent are unpatentable. IPR2019-00250 Patent 7,979,174 B2 3 A. Related Proceedings Petitioner identifies SZ DJI Technology Co. Ltd. v. Autel Robotics USA LLC, C.A. No. 16-706-LPS-CJB (D. Del.) and Certain Unmanned Aerial Vehicles and Components Thereof, Inv. No. 337-TA-113 (USITC, instituted Sept. 26, 2018) as related matters. Pet. 1. In addition to the matters identified by Petitioner, Patent Owner identifies a separate petition for inter partes review of the ’174 patent, concurrently filed with the above- captioned matter, and assigned case number IPR2019-00249. Paper 5 (Mandatory Notice), 2. Concurrently with this Decision, we issue a final written decision in IPR2019-00249, which addresses different claims of the ’174 patent. B. The ’174 Patent The ’174 patent relates to an autonomous vehicle including a control system having sensors that, in response to sensed conditions, adjusts the movement and speed of the vehicle. Ex. 1001, 2:1–10. Petitioner provides modified Figure 1 (Pet. 9), shown below, to illustrate the prominent features of the ’174 patent. IPR2019-00250 Patent 7,979,174 B2 4 Figure 1, above, is a block diagram of a system for providing automatic planning and regulation of the speed of autonomous vehicles which Petitioner has annotated to highlight the various elements and to indicate the output and input of “desired speed.” Ex. 1001, 2:14–16. In particular, system 100 includes one or more sensors 102 (green) for gathering data, which depend on the type of autonomous vehicle being used, and may include sensors collecting “data regarding speed limit zones, road conditions, tire pressure, etc.” Id. at 2:57–67. Speed planner 104 (light blue) and path planner 108 (yellow) receive input data from the sensors to calculate the desired speed for a planned path. Id. at 3:18–25. Speed controller 101 (purple) then generates a speed command that provides signals to one or more actuators 106 (red) to adjust the vehicle speed according to the desired speed. Id. at 7:4–10. Actuators, depending on the type of autonomous vehicle, may include “disc brakes, steering column, gas pedal, etc.” Id. at 7:12–14. The speed planner may also output a speed command category that “indicates why the desired speed was selected,” and IPR2019-00250 Patent 7,979,174 B2 5 a speed command category may be selected based on certain movement constraints. Id. at 8:24–44, 3:44–45. Certain constraints are given more weight (i.e., a “higher priority”) than others. Id. at 3:46–48. For example, collision avoidance has a higher priority than maintaining a certain speed. Id. C. Illustrative Claims Claims 9 and 14, claiming methods and programs to regulate or calculate the speed of an autonomous vehicle, are the only independent claims of the ’174 patent challenged in the instant proceeding. Claims 9 and 14 are reproduced below: 9. A method of regulating the speed of an autonomous vehicle, the method comprising: receiving an input from one or more sensors regarding conditions which affect the movement of the autonomous vehicle: calculating a desired speed based, at least in part, on the received input; and outputting the desired speed to a control system configured to use the output desired speed to calculate speed commands for use by one or more actuators to adjust the speed of the autonomous vehicle: wherein outputting the desired speed includes outputting a speed command category associated with the desired speed. Ex. 1001, 10:17–31. 14. A program product comprising program instructions embodied on a processor-readable medium for execution by a programmable processor, wherein the program instructions are operable to cause the programmable processor to: calculate a desired speed of an autonomous vehicle based on data received regarding conditions which affect movement of IPR2019-00250 Patent 7,979,174 B2 6 the autonomous vehicle and on one or more prioritized constraints; and output the calculated speed to a control system configured to use the output speed to calculate speed commands for use by one or more actuators to adjust the speed of the autonomous vehicle. Ex. 1001, 10:46–58. D. Instituted Grounds of Unpatentability We instituted trial to determine whether claims 9–17 of the ’174 patent would have been obvious1 in view of the following grounds of unpatentability: Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis 9–11 103(a) Andersson2, Urmson3 12–13 103(a) Andersson, Urmson, Fregene4 9–11 103(a) Andersson, Urmson, Ahmed-Zaid5 1 The Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29, 125 Stat. 284, 287–88 (2011), revised 35 U.S.C. § 103, effective March 16, 2013. Because the application from which the ’174 patent issued was filed before this date, the pre-AIA version of § 103 applies. 2 Andersson et al., US 6,836,719 B2, issued December 28, 2004 (“Andersson”) (Ex. 1004). 3 Urmson et al., A Robust Approach to High-Speed Navigation for Unrehearsed Desert Terrain, 23(8) J. Field Robotics 467 (2006) (“Urmson”) (Ex. 1008). 4 Fregene et al., US 2007/0078600 A1, published April 5, 2007 (“Fregene”) (Ex. 1006). 5 Ahmed-Zaid et al., US 6,968,266 B2, issued November 22, 2005 (“Ahmed- Zaid”) (Ex. 1007). IPR2019-00250 Patent 7,979,174 B2 7 Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis 12–13 103(a) Andersson, Urmson, Ahmed-Zaid, Fregene 14–17 103(a) Urmson, Augenbraun6, Fregene 15 103(a) Urmson, Augenbraun, Fregene, Mandow7 Dec. 7, 24. II. ANALYSIS A. Level of Ordinary Skill in the Art Relying on the testimony of Dr. Ducharme, Petitioner contends that one of ordinary skill in the art at the time of invention “would have had at least an undergraduate degree in robotics or computer engineering, or equivalent knowledge, training, or experience, with at least two years of experience working with the design and development of autonomous devices or speed control systems for autonomous speed adjustment.” Pet. 16; Ex. 1003 ¶¶ 19–20 (same). Although Patent Owner acknowledges that claims are interpreted from the perspective of one having ordinary skill in the art (PO Resp. 7), Patent Owner does not argue for any particular level of ordinary skill in the art in the Patent Owner Response. See generally PO Resp. Dr. Reinholtz, however, testifies that “a person of ordinary skill in the art (‘POSA’) at the 6 Augenbraun et al., US 2007/0061040 A1, published March 15, 2007 (“Augenbraun”) (Ex. 1005). 7 Mandow et al., Dynamic Speed Planning for Safe Navigation, 1 Proceedings of the 1997 IEEE/RSJ Int’l Conf. on Intelligent Robot and Sys. Innovative Robotics for Real-World Applications IROS 97, 231 (Sept. 1997) (“Mandow”) (Ex. 1009). IPR2019-00250 Patent 7,979,174 B2 8 time of the invention of the ’174 patent would have a Bachelor’s degree in engineering or computer science and would have had at least two years of experience developing control systems for electromechanical systems, including autonomous vehicles,” and that “[a]dditional education could substitute for professional experience and significant work experience could substitute for formal education.” Ex. 2001. ¶ 27. Upon review of the arguments set forth by Drs. Ducharme and Reinholtz, we adopt Dr. Reinholtz’s description of one of ordinary skill in the art, as it is consistent with the prior art of record and persuasively considers how the necessary level and type of education may be offset by work experience in the relevant field. We observe that neither party contends that the difference in education and experience level offered by Drs. Ducharme and Reinholtz would have any impact on the parties’ arguments or conclusions in this case. We agree with the parties and note that our conclusions in this Decision do not depend on which proposed definition is applied. B. Claim Construction In this inter partes review, claim terms in an unexpired patent are construed according to their broadest reasonable interpretation in light of the specification of the patent in which they appear. 37 C.F.R. § 42.100(b) (2018).8 In determining the broadest reasonable construction, we presume 8 A recent amendment to this rule does not apply here, because the Petition was filed before November 13, 2018. See Changes to the Claim Construction Standard for Interpreting Claims in Trial Proceedings Before the Patent Trial and Appeal Board, 83 Fed. Reg. 51,340 (Oct. 11, 2018) (amending 37 C.F.R. § 42.100(b) effective November 13, 2018) (now codified at 37 C.F.R. § 42.100(b) (2019)). At the time of the filing of the IPR2019-00250 Patent 7,979,174 B2 9 that claim terms carry their ordinary and customary meaning. See In re Translogic Tech., Inc., 504 F.3d 1249, 1257 (Fed. Cir. 2007). A patentee may define a claim term in a manner that differs from its ordinary meaning; however, any special definitions must be set forth in the specification with reasonable clarity, deliberateness, and precision. See In re Paulsen, 30 F.3d 1475, 1480 (Fed. Cir. 1994). In their briefing, the parties dispute the meaning of two claim phrases—(1) speed commands and (2) speed command category. See Pet. 16–19; PO Resp. 7–12. Patent Owner additionally identifies “minimum safe speed” and “constraints” as terms for construction. PO Resp. 12–15. Petitioner proposes that “speed command category” means “information output by the speed planner that ‘indicates why the desired speed was selected’” and that “speed commands” means “commands or signals to adjust the vehicle’s speed.” Pet. 16–17, 19. Patent Owner proposes the following constructions: “speed commands” to mean “commands relating to the speed of the autonomous vehicle”; “speed command category” to mean “information reflecting the reason for the desired speed”; “minimum safe speed” to mean “desired speed for the vehicle when obstacles are present”; and “constraint(s)” to mean “a limit that generally cannot be violated.” PO Resp. 8–15. Petition in this proceeding, the applicable claim construction standard was set forth in 37 C.F.R. § 42.100(b) (2018). IPR2019-00250 Patent 7,979,174 B2 10 We discern little difference between the parties’ respective constructions. Further, during the hearing both parties expressed a general view that claim construction was unlikely to affect the parties’ positions.9 Hr. 8:11–13, 9:13–15, 14:11–12, 15:2–4, 28:18–19, 43:3–12, 45:20–22; see also Reply 2 (asserting that “the challenged claims are unpatentable even under Patent Owner’s constructions”). Only terms that are in controversy need to be construed, and then only to the extent necessary to resolve the controversy. Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (applying Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999) in the context of an inter partes review). We determine none of the identified claim terms require construction to resolve the issues in dispute in this proceeding. C. Principles of Law To prevail in its challenges to the patentability of the claims, Petitioner must demonstrate by a preponderance of the evidence that the challenged claims are unpatentable. 35 U.S.C. § 316(e); 37 C.F.R. § 42.1(d). “In an [inter partes review], the petitioner has the burden from the onset to show with particularity why the patent it challenges is unpatentable.” Harmonic Inc. v. Avid Tech., Inc., 815 F.3d 1356, 1363 (Fed. Cir. 2016) (citing 35 U.S.C. § 312(a)(3) (requiring inter partes review petitions to identify “with particularity . . . the evidence that supports the 9 The parties disagree as to what is required for the prior art to teach an “autonomous vehicle,” but notably, neither party requested construction of that term in their respective briefs. See generally Pet.; PO Resp. We address the interpretation of the term “autonomous vehicle” below to the extent necessary to resolve the dispute raised by the parties’ substantive arguments. IPR2019-00250 Patent 7,979,174 B2 11 grounds for the challenge to each claim”)). This burden of persuasion never shifts to Patent Owner. See Dynamic Drinkware, LLC v. Nat’l Graphics, Inc., 800 F.3d 1375, 1378 (Fed. Cir. 2015) (discussing the burden of proof in inter partes review). A patent claim is unpatentable under 35 U.S.C. § 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. See 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 ordinary skill in the art; and (4) if in the record, objective evidence of nonobviousness. Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966). Consideration of the Graham factors “helps inform the ultimate obviousness determination.” Apple Inc. v. Samsung Elecs. Co., 839 F.3d 1034, 1048 (Fed. Cir. 2016) (en banc), cert. denied, 138 S. Ct. 420 (2017). To prevail in an inter partes review, Petitioner must explain how the proposed combinations of prior art would have rendered the challenged claims unpatentable. At this final stage, we determine whether a preponderance of the evidence of record shows that the challenged claims would have been rendered obvious in view of the asserted prior art. We analyze the asserted grounds of unpatentability in accordance with these principles. IPR2019-00250 Patent 7,979,174 B2 12 D. Obviousness of Claims 9–11 in view of Andersson and Urmson alone and in combination with Ahmed-Zaid Petitioner contends the subject matter of claims 9–11 would have been obvious over the combined disclosures of Andersson, Urmson, alone or in further combination with Ahmed-Zaid. Pet. 19, 47. 1. Andersson Andersson discloses systems and methods for manually or autonomously controlling vehicle speed. See Ex. 1004, 2:32–34, 3:15–18, 4:29–30. Andersson explains: [T]he method may also comprise the steps of extracting, from a plurality of sensors, sensor information regarding obstacles on or in [the] vicinity of said road section, and generating, based on information from the input connection, the map database, the driver information database, and the sensor information, a speed plan for the current and/or upcoming road section. Id. at 2:41–47. The speed plan is generated in speed plan generator 12. Id. at 4:24–26. The speed plan is sent to a control device that, in turn, controls brake control unit 9 and throttle control unit 10, in accordance with the speed plan. Id. at 3:16–19, 3:58–4:2. Andersson also includes a route planner that enables the speed planner to generate “a speed plan for the entire route . . . , thereby estimating mean speed, travel time, fuel consumption and so on.” Id. at 6:26–30. Figure 2, reproduced below, is illustrative of one embodiment of Andersson. IPR2019-00250 Patent 7,979,174 B2 13 Figure 2 above is a “block diagram disclosing certain preferred additional functions of the block diagram in accordance with [Figure 1].” Id. at 3:27– 29. The system of Figure 2 includes controller 1, computer 2, and speed plan generator 12. Id. at 3:33–4:28. Information from map database 3, driver preference database 4, and input port 5 (which receives positioning signals 11) is used to generate a speed plan that is transmitted to vehicle control system 8. Id. at 4:6–34. Additionally, sensor input from environmental sensors 15 maybe included as input to speed plan generator 12. Id. at 5:43–47. Environmental sensors 15 may include “radar sensors, [infrared (“IR”)] sensors or a camera” and can be used to detect “light conditions, visibility, and road friction.” Id. at 5:47–50. System controller 8 is “connected with [sic] brake control unit 9 and/or an engine control unit such as throttle control unit 10” in order to enable autonomous control of the vehicle. Id. at 3:15–19, 3:58–4:2. 2. Urmson Urmson describes autonomous vehicles, their hardware and software, and modes of operation. See Ex. 1008, 468–469. The vehicles of Urmson IPR2019-00250 Patent 7,979,174 B2 14 are equipped with sensors, such as light detecting and ranging (“LIDAR”) and radio detecting and ranging (“RADAR”) systems and LICK laser measurement sensors (“LMS”) to sense terrain and vehicle position. Id. at 476–479. The Urmson vehicles have “[o]nboard navigation software [that] combines incoming sensor data with a preplanned route to generate a new safe and traversable route.” Id. at 479, Fig. 9. These sensors also send desired speed information to a speed controller which operates the throttle and brakes to adjust speed accordingly. Id. at 475. Urmson includes a speed planner that “operates on the output of the geometric planner and pre- emptively slows the robot for any sharp turns that may result when the geometric planner generates a plan to avoid obstacles.” Id. at 488. 3. Ahmed-Zaid Ahmed-Zaid describes an adaptive cruise control system to control the speed of a vehicle. See Ex. 1007, 1:15–19, 3:24–32. Ahmed-Zaid’s system includes a controller that “generates a predicted future path profile” “in response to the object profile [from detection system 28] and the navigation signal,” and “inhibits resume speed of the automotive vehicle in response to the predicted future path profile.” Id. at 2:35–40. Each of the “object profile, the yaw rate, the speed of the vehicle 22, and the navigation signal determines an operating mode of the system 20.” Id. at 4:14–16. Ahmed- Zaid describes several operational modes, including “follow mode, cruise mode, auto resume mode, or inhibit resume mode,” that adjust the speed of the vehicle based on certain conditions. Id. at 4:17–18. The “follow mode,” for example, continuously adjusts the vehicle speed so that the vehicle maintains a certain distance from a vehicle traveling ahead. Id. at 4:18–22. In an auto resume mode or inhibit resume mode, the vehicle either resumes IPR2019-00250 Patent 7,979,174 B2 15 acceleration or prevents acceleration based on sensor information that indicates whether an object is detected in the future path of the vehicle. Id. at 2:28–58, 4:24–26. 4. Analysis of Claim 9 Petitioner contends that independent claim 9 would have been obvious over the combination of Andersson and Urmson. Pet. 19. Alternatively, Petitioner contends that claim 9 would have been obvious over the combination of Andersson, Urmson, and Ahmed-Zaid. Pet. 47. Petitioner’s alternative position that Ahmed-Zaid, in addition to Andersson, describes a “speed command category” in the event Patent Owner argues the term requires both why and how the actuators adjust the vehicles speed.” Pet. 36, n.11, 47–50. Neither party proposes such a construction. PO Resp. 29. Therefore, we do not address this alternative position. Claim 9 is directed to a method of regulating the speed of an autonomous vehicle that includes (1) receiving input from one or more sensors that obtain data that affects the movement of the vehicle; (2) calculating the desired speed based on data from the sensors; (3) outputting the desired speed to a control system that calculates speed commands based on the desired speed; (4) one or more actuators that adjust the speed of the vehicle based on speed commands received from the control system; and (5) where outputting the desired speed includes outputting a speed command category associated with the desired speed. Ex. 1001, 10:17–30. Relevant to claim 9, Petitioner contends that Andersson describes an autonomous vehicle.10 Pet. 19–20 (citing Ex. 1004, 2:32–34, 3:15–19, 3:67– 10 The term “autonomous vehicle” appears both in the preamble and the body of claim 9. See Ex. 1001, 10:16–29. We need not decide whether the IPR2019-00250 Patent 7,979,174 B2 16 4:1, 4:28–33).11 According to Petitioner, the vehicle of Andersson has a plurality of sensors, including a position sensor, a compass, steering angle sensor, yaw sensor, wheel speed sensor, radar sensors, IR sensors, or a camera. Id. at 20–22 (citing Ex. 1004, 2:40–42, 3:38, 5:43–51, 6:32–37, Fig. 2; Ex. 1003 ¶ 82). Petitioner argues that Andersson has a speed plan generator, corresponding to the claimed speed planner, that “‘extract[s], from a plurality of sensors, sensor information regarding obstacles on or in vicinity of said road section’ and then ‘generat[es], based on … the sensor information, a speed plan for the current and/or upcoming road section.’” Id. at 22–23 (quoting Ex. 1004, 2:40–47) (citing Ex. 1004, 2:64–66, 3:56– 60, 4:26–28; Ex. 1003 ¶ 86 & n.6).12 Petitioner explains, however, that Andersson “lacks specific details about how the sensor inputs are used to generate the desired speeds,” but “Urmson discloses just such specific equations/algorithms . . . for calculating a desired speed,” and “use[s] the same or similar inputs as described in Andersson.” Id. at 24–26. Petitioner explains that the speed planner of Urmson performs a “first pass [that] walks the path in the forward direction and sets a maximum speed at each point” on the path and then performs a “‘second pass [that] walks from the last point to the first point and limits the change in velocity so that it is constrained by [equation] (4).’” Id. at 25. Petitioner reasons that a person of ordinary skill in the art would have been preamble is limiting because Petitioner has shown that the recitation of “an autonomous vehicle” is satisfied by the prior art. 11 Petitioner also alleges that Urmson discloses “an autonomous vehicle” as claimed. Pet. 20 n.7. 12 Petitioner further contends Urmson discloses sensors, i.e., LIDAR and RADAR, used to detect obstacles. Pet. 22 n.8. IPR2019-00250 Patent 7,979,174 B2 17 “motivated to look to another reference disclosing details for calculating speeds, such as Urmson” in light of the teachings of Andersson. Id. at 26 (citing Ex. 1008, 490–91; Ex. 1003 ¶¶ 93–94). Petitioner further reasons that use of “Urmson’s algorithms would make Andersson’s system safer, more reliable, and more comfortable for passengers.” Id. at 27 (citing Ex. 1008, 490; Ex. 1003 ¶¶ 93–94). Petitioner also alleges that Andersson describes a vehicle control system (system controller 8) that receives the speed plan and calculates the speed commands, i.e., signals to adjust the vehicle’s speed, based on the speed plan. Id. at 28–30 (citing Ex. 1004, 3:57–4:2, Fig. 2; Ex. 1003 ¶¶ 98– 99). Petitioner acknowledges, however, that Andersson fails to describe the details of how system controller 8 calculates speed commands, but explains that Urmson similarly teaches a speed controller which is a piece of software that compares a “measured speed” against a “speed error” and then adjusts the engine or break controllers according to the commands received. Id. at 31–32 (citing Ex. 1008, 475, Fig. 9; Ex. 1003 ¶¶ 103–106). Petitioner explains that “[a person having ordinary skill in the art (‘PHOSITA’)] reviewing Andersson’s disclosure would thus be motivated to look to another reference disclosing specific details for calculating speed commands based on a desired speed, such as Urmson,” for the reasons stated above. Id. at 32 (relying on Petition Section VIII (A)(1)(b) and citing Ex. 1003 ¶ 107). Petitioner also asserts that Andersson describes a “brake control unit” and a “throttle control unit,” which correspond to the “one or more actuators” as claimed. Id. at 28, 30–31 (citing Ex. 1004, 3:64–4:2, 5:14–16, Fig. 2; Ex. 1003 ¶¶ 101, 112–114, 116, 167–168). Petitioner similarly argues that Urmson’s controller commands actuators to adjust the speed of IPR2019-00250 Patent 7,979,174 B2 18 the vehicle as claimed by the ’174 patent. Id. at 31–32 (citing Ex. 1008, 474–475, Fig. 9; Ex. 1003 ¶¶ 103–106, 115). Lastly, Petitioner urges that Andersson provides a speed command category providing information about why the desired speed is selected. Id. at 33. In particular, Petitioner draws attention to Andersson’s teachings that combine two modes of operation which “display[s] speed changes on the display unit 6, in order to inform the driver of up-coming speed changes.” Id. at 34 (quoting Ex. 1004, 4:2–5; citing Ex. 1004, 3:60–65, 5:14–20, Fig. 2). In one example, display unit 6 “reads ‘Slow down!,’ ‘Dangerous curve ahead,’ and ‘Recommended speed 25 mph.’” Id. at 34– 35 (quoting Ex. 1004, Fig. 2) (citing Ex. 1003 ¶ 123). Patent Owner’s arguments regarding claim 9 focus on three purportedly missing elements and the lack of a motivation to combine the references. See PO Resp. 25–33. After reviewing the parties’ briefing and evidence of record, we are persuaded by Petitioner’s arguments and supporting evidence for the undisputed limitations of claim 9 as summarized above. Accordingly, we determine that Petitioner has established by a preponderance of the evidence that each of the limitations of claim 9 not challenged by Patent Owner are present in the combination of Andersson and Urmson. See In re NuVasive, 841 F.3d 966, 974 (Fed. Cir. 2016) (explaining that the Board need not make specific findings as to claim limitations that Patent Owner does not dispute are disclosed in the prior art). We address Patent Owner’s specific arguments below. IPR2019-00250 Patent 7,979,174 B2 19 a. whether Andersson discloses an “autonomous vehicle” Patent Owner contends that Andersson does not disclose an autonomous vehicle but rather “relates to cruise control technologies and has a driver present at all times.” PO Resp. 25. Patent Owner asserts that the ’174 patent describes vehicles that are capable of maneuvering and navigating without the presence of a driver such that an autonomous vehicle, as understood by the ’174 patent, “control[s] both steering and speed.” Id. (citing Ex. 2001 ¶ 78) Patent Owner argues that because Andersson relates only to cruise control technology—and not to steering—it describes a very different vehicle than that claimed by the ’174 patent. Id. at 24–25. In weighing the evidence before us, we disagree with Patent Owner’s arguments. Based on our review of the arguments and evidence of record, we determine that the term “autonomous vehicle” may include varying levels of autonomy, and that although it may include both speed and steering control, it does not require the presence of both as Patent Owner urges (PO Resp. 25). This understanding is consistent with the Specification of the ’174 patent, which refers to autonomous vehicles embodied in the ’174 patent exclusively in the context of speed control. See Ex. 1001, 2:1–10, 2:40–4, 3:29–50, 3:58–4:2, 4:3–5, 4:29–33, 5:1–11, 5:51–55, 6:4–7, 6:15– 20, 8:23–27, 8:65–9:7, 9:18–22, Figs. 1, 4. For example, the ’174 patent explains that [c]onventional path planning algorithms, however, do not inherently indicate the speed at which the vehicle should traverse the path. By automatically providing appropriate speed targets that are safe, feasible and achievable by the autonomous vehicle during path traversal, embodiments of the present invention IPR2019-00250 Patent 7,979,174 B2 20 enable an autonomous vehicle to complete tasks safely and quickly. Id. at 2:43–49 (emphasis added). According to Patent Owner, the statement in the ’174 patent that an autonomous vehicles is “capable of navigating and maneuvering without human control,” “needs to determine the characteristics of its environment required to enable it to carry out the task it has been assigned,” and “needs to plan and follow a path to its destination while detecting and avoiding obstacles,” supports its contention that autonomous vehicles require both speed and steering control. Id. at 1:53–61; PO Resp. 25; Sur-reply 4; Ex. 2001 ¶¶ 77–78. However, this lone statement, describing the goals of the Defense Advanced Research Projects Agency (“DARPA”) Grand Challenge—not the embodiments of the ’174 patent—is insufficient to convey to the skilled artisan that the term autonomous vehicle in the context of the ’174 patent requires only full autonomy (i.e., a system comprising both speed and steering control). Because we find that the claimed autonomous vehicle does not require full autonomy, we determine that Andersson’s teaching of an adaptive cruise control system based on both sensor input and database information is sufficient to teach an autonomous vehicle.13 Therefore, having considered the record evidence and arguments advanced, we determine that Petitioner has demonstrated by a 13 As we observed in our Decision on Institution (Dec. 12 n.4), Petitioner argues that Urmson, in addition to Andersson, describes an “autonomous vehicle.” Pet. 20, n.7 (citing Ex. 1008, 468–469); Ex. 1003 ¶ 78, n.4 (same). Patent Owner does not address Petitioner’s assertion. See generally PO Resp. Rather, Patent Owner, as well as Dr. Reinholtz, describes Urmson as an autonomous vehicle. Pet. 19; Ex. 2001 ¶¶ 65–68. IPR2019-00250 Patent 7,979,174 B2 21 preponderance of the evidence that an “autonomous vehicle” is taught or suggested by Andersson. b. whether the combination of Andersson and Urmson teaches or suggests a “speed planner” Patent Owner also asserts that “neither Andersson nor Urmson disclose or render obvious element 9[b], which requires a speed planner that ‘calculates a desired speed based, at least in part, on the received sensor input.’” PO Resp. 25. According to Patent Owner, Petitioner’s argument that the person skilled in the art “would have understood that generating a speed plan based on sensor information would result in Andersson’s speed plan generator performing at least one calculation” must fail because “Andersson’s selection of user defined speeds requires no calculation” and because “Petitioner’s position falls far short of the standard for inherency, which requires inevitability.” Id. at 26. Patent Owner also argues that “[b]ecause Urmson calculates speeds in advance, its calculations are not based on sensor input as recited in element 9[b].” Id. at 27. We are persuaded by Petitioner’s evidence and argument. Andersson teaches that its speed plan generator is responsible for generating or calculating the speed plan transmitted to the vehicle system controller and is responsible for engaging the brake control unit or the engine control unit. Ex. 1004, 2:59, 3:15–19, 3:54–4:34, 6:28–30. Andersson describes a combination of inputs including information from environmental sensors, a map database, a driver information database, and positioning systems as information fed into the speed plan generator. Id. at 1:8–10, 2:42–46, 2:52– 53, Fig. 2. Andersson explains that the sensor data, which may include “information regarding obstacles on or in the vicinity of said road section,” IPR2019-00250 Patent 7,979,174 B2 22 may be used in combination with other input data to generate a speed plan for “the current and/or upcoming road section[s].” Id. at 2:41–47. As a result, the speed plan generated by the speed plan generator may account for “temporary circumstances or objects not present in the map database.”14 Id. at 2:47–49. Dr. Ducharme testifies that Andersson’s speed plan generator must perform one or more calculations to interpret the data from the sensors (e.g., data regarding current vehicle position and speed) in order to calculate a desired speed (e.g., a speed that is 5 mph less than the current speed). Indeed, even simple adjustments, such as increasing the vehicle’s speed to a higher preferred speed or decreasing the vehicle’s speed to a lower preferred speed, necessarily require[s] calculations (i.e., addition and subtraction, respectively). Ex. 1003 ¶ 88. We credit Dr. Ducharme’s unrebutted testimony in this regard. We acknowledge Dr. Reinholtz’s testimony that Andersson does not calculate speed because Andersson generates speed using a “neural network that relates said plurality of map information with said plurality of driver behavior information.” Ex. 2001 ¶ 85 (emphasis added). Even assuming arguendo that a neural network—i.e., a network that “attempt[s] to emulate the human decision-making process by learning to recognize patterns and 14 Dr. Reinholtz also testifies that “Andersson at most discloses the use of sensor data to select a user defined speed,” and therefore does not use sensor data to calculate the desired speed. Ex. 2001 ¶ 84 (citing Ex. 1004, 5:44– 55). Dr. Reinholtz’s testimony however is in relation to use of sensor data relevant to “factors not directly related to the upcoming road section, but rather to a driver-vehicle environmental combination” (Ex. 1004, 5:44–48) and does not account for the use of sensor data to account for temporary circumstances or unexpected objects in the vehicle path (id. at 2:41–49). IPR2019-00250 Patent 7,979,174 B2 23 make decisions in a humanlike way” (id.)—does not involve performing calculations as Patent Owner alleges,15 the neural network of Andersson is part of the computing device and not the speed plan generator responsible for generating the speed plan. Ex. 1004, 3:9–11 (“[T]he computing device is suitably realized by means of a neural network.”), 4:47–49 (explaining that in a preferred embodiment, “the computing device 2 . . . comprises a neural network unit”). Dr. Reinholtz further testifies that that there is no disclosure in Andersson to suggest whether the speed plan generator also includes a neural network. Ex. 1027, 23:12–24:4 (referring to Ex. 1025, 428:21– 429:10). Accordingly, Patent Owner’s suggestion that the speed planner does not calculate—but rather relates information—because it uses a neural network, is not borne out by the evidence of record and does not negate our finding that Andersson’s speed plan generator calculates a desired speed. As discussed above, Petitioner relies on Urmson for teaching specific details of calculating the desired speed. With respect to Petitioner’s reliance on Urmson’s teachings, we are similarly persuaded that Urmson suggests a speed planner as claimed. Like Andersson, the speed planner of Urmson “is responsible for ensuring driving speeds are safe.” Ex. 1008, 490. Urmson uses onboard sensors to avoid and account for obstacles on the planned route. Id. at 469, 476 (“LIDAR and the RADAR are used to detect obvious obstacles at long ranges.”), 478. Navigating obstacles is identified as a principal risk because “[h]itting large obvious obstacles . . . can destroy a vehicle.” Id. at 481. Sensor information is fed into the geometric planner 15 There is evidence of record to suggest that a neural network may involve calculations, in particular by applying “the weighted sum of inputs.” Ex. 2005; Ex. 1027, 45:21–46:5. IPR2019-00250 Patent 7,979,174 B2 24 and then to the speed planner. Id. at 480 (Fig. 14). According to Urmson, in “[t]he first stage, the geometric planner, adjusts the path to avoid obstacles . . . [and then] [t]he speed planner operates on the output of the geometric planner and pre-emptively slows the robot for any sharp turns that may result when the geometric planner generates a plan to avoid obstacles.” Id. at 488; see also id. at 490 (explaining that the “speed planner will slow the output path to account for obstacle avoidance”). Urmson makes clear that its speed plan is not based only on calculations occurring in advance (PO Resp. 26–27), but rather, Urmson’s speed planner combines both preplanned and real-time information to generate a new speed plan. Ex. 1008, 479 (“[I]ncoming sensor data [is combined] with a preplanned route to generate a new safe and traversable route” (emphasis added)). Further, Patent Owner’s additional argument that “Urmson’s ‘Speed Planner’ operates on the output of Urmson’s ‘Geometric Planner’ [and therefore] Urmson’s sensors do not input to the ‘Speed Planner’” lack merit. Sur-reply 9. Patent Owner’s claims do not require direct input of sensor information into the speed planner; rather, the claims require only that the speed planner calculate a desired speed based, in part, on data obtained from one or more sensors. Accordingly, the weight of the evidence supports a determination that Petitioner has established by a preponderance of the evidence that the combination of Andersson and Urmson teaches “a speed planner . . . configured to calculate a desired speed based, at least in part, on data obtained from the one or more sensors,” as claimed. IPR2019-00250 Patent 7,979,174 B2 25 c. whether Andersson teaches or suggests a “speed command category” In addition to the arguments discussed above, Patent Owner asserts that “none of Andersson, Urmson, and/or Ahmed-Zaid disclose or render obvious [the] . . . recited ‘speed command category.’” PO Resp. 27. According to Patent Owner, Andersson does not output a speed command category as part of the speed plan, but rather, “Andersson explicitly defines the speed planner output (i.e., the speed plan) as ‘speed, accelerations and retardations” and nothing in Andersson suggests that the messages displayed on display unit 6 (illustrated in Figures 1 and 2) are output from the speed planner. Id. at 28 (citing Ex. 1004, 3:58–4:5). We are persuaded by Petitioner’s argument and evidence. Figure 2 of Andersson, reproduced below, illustrates the relevant features relating to the “speed command category.” Andersson’s Figure 2, above, “is a block diagram disclosing certain preferred additional functions” in accordance with one preferred IPR2019-00250 Patent 7,979,174 B2 26 embodiment. Ex. 1004, 2:40–49, 3:25–29. In the embodiment of Figure 2, both sensor data from environmental sensors 15 a driver behavior model output from computing device 2 is input into speed plan generator 12 to generate a vehicle plan for a current and up-coming road section. Id. at 3:50–57. “The speed plan, containing information regarding preferred vehicle speed, accelerations and retardations . . . is thereafter inputted to a mode selector 7,” where the user chooses between a manual drive mode, in which speed information and recommendations are displayed on unit 6, or an autonomous control mode where the speed plan is input into system control device 8 to control braking and acceleration. Id. at 3:58–67. Andersson explains that it is possible to use both operational modes simultaneously, such that the upcoming changes may be viewed on display 6 in an autonomous control mode. Id. at 4:2–5; see also id. at 5:14–20 (“[I]n autonomous speed control mode . . . an indication of a speed limit condition and vehicle control actions may be displayed to the driver . . . .”). The speed plan also includes “information regarding [the] appropriate changes of the vehicle speed in order to respond to said driver model,” which is “transmitted to the manual/automatic mode selector 7, as described above, and the signal is thereafter transmitted to the vehicle system control device 8 for autonomous speed adaptation and/or the display unit 6, for displaying appropriate speed alterations to the driver of the vehicle.” Id. at 4:25–34. Therefore, the speed plan of Andersson includes more than simply vehicle speed, accelerations, and retardations, as Patent Owner alleges. See id. 2:35–39. IPR2019-00250 Patent 7,979,174 B2 27 Therefore, we determine that Petitioner has demonstrated by a preponderance of the evidence that a “speed command category” is taught or suggested by Andersson. d. whether a person ordinarily skilled in the art would have had reason to combine Andersson and Urmson Patent Owner contends that the grounds based on Andersson and Urmson (with or without Ahmed-Zaid) “also fail because there is no motivation to combine the prior art” as Petitioner suggests. PO Resp. 30. First, Patent Owner argues that Andersson teaches away from speed calculation and speed command categories because Andersson “selects speeds using a neural network, which is not calculating speed to a POSA.” Id. Further, Patent Owner contends that “Andersson’s disclosure . . . is limited to the display of information that is not based on sensor input, and a POSA would not be motivated to modify Andersson to calculate speeds or display information based on sensor input, including speed command categories, because Andersson does not disclose, teach, or suggest calculation of desired speeds based on sensor input.” Id. at 30–31. Therefore, according to Patent Owner, Andersson teaches away from modification. Id. We do not agree with Patent Owner’s argument that Andersson teaches away.16 To teach away, a reference must discourage one of ordinary skill in the art from following the path set out in the reference, or lead that 16 As an initial matter, Patent Owner’s “teaching away” argument simply repeat the arguments presented (and rejected) above in Section II.D.4.b. and c., i.e., Patent Owner’s position that the prior art does not teach a “speed planner” or a “speed command category,” as claimed. IPR2019-00250 Patent 7,979,174 B2 28 person in a direction divergent from the path taken by the applicant. In re Gurley, 27 F.3d 551, 553 (Fed. Cir. 1994)(“[A] reference will teach away if it suggests that the line of development flowing from the reference’s disclosure is unlikely to be productive of the result sought by the applicant.”). “A reference does not teach away . . . if it merely expresses a general preference for an alternative invention but does not ‘criticize, discredit, or otherwise discourage’ investigation into the invention claimed.” DePuy Spine, Inc. v. Medtronic Sofamor Danek, Inc., 567 F.3d 1314, 1327 (Fed. Cir. 2009) (quoting In re Fulton, 391 F.3d 1195, 1201 (Fed. Cir. 2004)). “[S]ilence does not imply teaching away.” Allergan, Inc. v. Apotex Inc., 754 F.3d 952, 964 (Fed. Cir. 2014). Therefore, the fact that Andersson exemplifies a neural network—in a preferred embodiment (Ex. 1004, 4:47– 60, 6:7–9)—does not teach away from pursuing the path taken in the ’174 patent. Similarly, Patent Owner’s additional position that because the speed planner of Andersson does not calculate speeds or output speed command categories based on sensor input, it teaches away from modifying Andersson’s system, is unsound. Silence, without more, is not a teaching away. Allergan, 754 F.3d at 964. Next, Patent Owner argues that because Andersson and Urmson address fundamentally different problems, a person of ordinary skill in the art would not have had reason to combine their teachings. PO Resp. 31. According to Patent Owner, “Andersson relates to cruise control technologies and determining speed along route segments,” and Urmson “relates to autonomous vehicles where route planning leads to path planning, and both speed and heading (steering angle) must be controlled in coordination.” Id. Patent Owner explains that because Andersson does not IPR2019-00250 Patent 7,979,174 B2 29 discuss path planning, “Andersson does not relate to autonomous vehicles and is a very different technology from Urmson.” Id. On this record, we disagree with Patent Owner’s arguments. Whether Andersson’s teachings are limited to adaptive cruise control systems, as Patent Owner alleges, is not the inquiry.17 Perfect overlap in technologies is not required. Rather, one of ordinary skill can use his or her ordinary skill, creativity, and common sense to make the necessary adjustments and further modifications to result in a properly functioning device. See KSR, 550 U.S. at 418 (“[A] court can take account of the inferences and creative steps that a person of ordinary skill in the art would employ”). Furthermore, “if 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.” See id. at 417. Here, Andersson’s methods and systems, at a minimum, control vehicle speed. Ex. 1004, 1:57–59 (“By such a method, it is possible to generate a speed plan, which may be used to adapt the speed of the vehicle for a chosen road section in a suitable manner.”), 2:25–32 (“[T]he method comprises the step of feeding said outputted speed plan to a throttle control device for controlling the throttle of a vehicle . . . .”), 3:58–4:1 (explaining that the system control device is “connected wit[h] a brake control unit 9 and/or an engine control unit such as a throttle control unit 10, for autonomous vehicle speed control.”). Urmson, similarly, relates to speed control in addition to steering control. Ex. 1008, 474 (“[S]peed control must 17 Patent Owner’s argument, based on its assertion that Andersson does not describe an autonomous vehicle, was rejected above in Section II.D.4.a. IPR2019-00250 Patent 7,979,174 B2 30 be accurate and responsive as it is constantly being adjusted to ensure vehicle stability.”), 491 (“[T]he algorithm runs on the output of the planner . . . allowing the vehicle to slow for unexpected obstacles.”). Petitioner relies on Urmson to describe “how the sensor inputs are used to generate the desired speeds.” Pet. 26 (“[A] PHOSITA . . . would have been motivated to look to . . . Urmson, to find the details for generating the desired speeds, including specific algorithms used by a speed planner to calculate [the] desired speeds.”). Therefore, overlap exists in the teachings of Andersson and Urmson in the aspect of speed control—the very aspect that is improved by modifying Andersson to include Urmson’ algorithms in the combination advanced by Petitioner. Lastly, Patent Owner argues that Urmson teaches away from a combination with Andersson “because any combination would only add complexity without improving functionality.” PO Resp. 31. Here, Patent Owner relies on the testimony of Dr. Reinholtz who explains that even minor alterations to the programming algorithm would result in “wide ranging impacts on other aspects of the system.” Id. (citing Ex. 2001 ¶¶ 119–121). According to Patent Owner, because “Urmson focuses on the simplicity of the interaction between its Navigation Subsystem and Vehicle Subsystem” and “[c]alculating desired speeds based on sensor input and outputting speed command category would only add complexity to Urmson’s system,” Urmson thus teaches away. Id. at 32 (citing Ex. 1008, sec. 2.2). Again, we disagree with Patent Owner’s teaching away argument. As explained above, teaching away requires that the reference “‘criticize, discredit, or otherwise discourage’ investigation into the invention claimed.” IPR2019-00250 Patent 7,979,174 B2 31 DePuy Spine, 567 F.3d at 1327. In this case, Patent Owner identifies no such criticism in Urmson (and we find none) that would have cautioned the skilled artisan against pursuing the solution described in the ’174 patent by combining aspects of Urmson with Andersson. See generally PO Resp.; Sur-reply. And while Urmson does discuss the simplicity of its design, “[t]he fact that the motivating benefit comes at the expense of another benefit, however, should not nullify its use as a basis to modify the disclosure of one reference with the teachings of another. Instead, the benefits, both lost and gained, should be weighed against one another.” Winner Int'l Royalty Corp. v. Wang, 202 F.3d 1340, 1349 n.8 (Fed. Cir. 2000). Furthermore, Petitioner does not propose to modify Urmson as Patent Owner suggests. Pet. 31–32. Instead, Petitioner proposes modifying Andersson with algorithms taught by Urmson. Accordingly, no complexity is added to Urmson. Therefore, we determine that Petitioner has demonstrated by a preponderance of the evidence that each limitation of claim 9 is taught or suggested by the combination of Andersson and Urmson and further that the skilled artisan would have had reason to make the suggested combination with a reasonable expectation of success. 5. Analysis of Claims 10 and 11 With respect to dependent claims 10 and 11, Petitioner identifies where the combination of Andersson and Urmson suggests each limitation of these claims. Pet. 37–41. Petitioner demonstrates that Urmson describes a geometric planner that calculates a planned path and the speed planner determines the maximum safe speed for points on the planned path, as required by claims 10 and 11. Id. at 37–40 (citing Ex. 1008, 479, 488, 490– IPR2019-00250 Patent 7,979,174 B2 32 91; Figs. 14, 25; Ex. 1003 ¶¶ 132–136, 172–174; Ex. 1004, 6:27–31). With respect to claim 11, Petitioner also demonstrates that “Andersson discloses that ‘it may be possible to connect a system as stated above with a route planner’ and ‘[b]y knowing the intended route, a speed plan for the entire route may be calculated in advance, thereby estimating mean speed, travel time, fuel consumption and so on.’” Id. at 40–41 (quoting Ex. 1004 at 6:27– 31). Further, Petitioner explains that Urmson also discloses that its speed algorithms “process[] the entire path once setting speeds well beyond the sensing horizon of the vehicle.” Pet. 41, n.12 (quoting Ex. 1008, 491). Thus, according to Petitioner, Andersson and Urmson disclose the additional limitations of claim 11. Patent Owner does not substantively address Petitioner’s arguments with respect to claims 10 and 11 in its Patent Owner Response. See generally PO Resp. Instead, Patent Owner argues that “[c]laims 10 and 11 and [sic] are patentable for the same reasons set forth above with regard to claim 9.” Id. at 33. Upon review of the arguments and evidence of record, we determine that Petitioner has demonstrated by a preponderance of the evidence that each of the limitations recited in claims 10 and 11 are present in or suggested by the combination of Andersson and Urmson and the skilled artisan would have had reason to make the combination with a reasonable expectation of success. See NuVasive, 841 F.3d at 974. E. Obviousness of Claims 12 and 13 in view of Andersson, Urmson, and Fregene, alone or in combination with Ahmed-Zaid Petitioner contends the subject matter of claims 12 and 13 would have been obvious over the combined disclosures of Andersson, Urmson, and IPR2019-00250 Patent 7,979,174 B2 33 Fregene, alone or in combination with Ahmed-Zaid. Pet. 41, 53. As discussed above, we do not address the parties’ contentions regarding Petitioner’s alternative position based additionally on Ahmed-Zaid. 1. Fregene Fregene describes methods and systems for collision avoidance for autonomous vehicles. Ex. 1006 ¶¶ 7–10, 22. The disclosed collision avoidance system utilizes two types of constraints. Id. ¶ 38. “The first constraint type includes the constraints under normal operation of the vehicle, . . . called the nominal constraints. The second constraint type includes the constraints under an emergency operation of the vehicle, . . . called the emergency constraints.” Id. “The nominal constraints are more restrictive than the actual operating limits of the vehicle, since maximum performance is used for a vehicle only in emergency situations.” Id. Emergency constraints “include maximum vehicle acceleration, minimum vehicle acceleration, maximum vehicle speed and minimum vehicle speed.” Id. ¶ 25. When normal collision avoidance schemes fail, vehicles apply emergency maneuver procedures. Id. ¶¶ 53–57. When the emergency has passed, “[t]he on-board processor determines what instruction the vehicle requires to switch out of the emergency maneuver mode and back to normal operation mode.” Id. ¶ 64. 2. Analysis of Claims 12 and 13 Claim 12 depends from claim 9 and additionally requires that calculating the desired speed is “based on one or more prioritized constraints such that the autonomous vehicle does not violate the one or more prioritized constraints.” Ex. 1001, 10:38–41. Claim 13 depends from claim 12 and further requires that “calculating the desired speed based on one or IPR2019-00250 Patent 7,979,174 B2 34 more prioritized constraints includes permitting violation of a lower priority constraint in order to avoid violating a higher priority constraint.” Id. at 10:42–45. Petitioner contends that the combination of Andersson, Urmson, and Fregene renders claim 12 obvious. Pet. 41. Petitioner argues that both Urmson and Fregene suggest the additional limitation of claim 12. Id. at 41– 43. In particular, Petitioner alleges that Urmson teaches or suggests this limitation. Id. at 41 (referring to Section XII(A)(1)(a)(ii) of the Petition). According to Petitioner, Urmson describes at least three scenarios where the desired speed is based on prioritized constraints that are not violated. Pet. 58–61. Petitioner first explains that Urmson’s navigation software is programmed to calculate the desired speed based on one or more prioritized constraints. Id. at 58. According to Petitioner, “Urmson’s speed planner ‘is responsible for ensuring driving speeds are safe’ and ‘slows [the autonomous vehicle] as necessary to swerve around an approaching obstacle.’ Thus, similar to the speed constraint in the ’174 patent, Urmson’s navigation software prioritizes ‘safe’ speeds as a constraint for determining a desired vehicle speed.” Pet. 59 (alterations in original) (quoting Ex. 1008, 490, 491; Ex. 1003 ¶ 274). “Therefore, Urmson’s speed planner is programmed to calculate a desired speed (i.e., a reduced speed) based on the high priority constraint of collision avoidance.” Id. (citing Ex. 1008, 490, 491; Ex. 1003 ¶ 274). Next, Petitioner asserts that Urmson describes a “two- pass process” where the maximum safe speed is determined for each point on the path—the first pass constrains the lateral velocity of the vehicle and the second pass limits rate of change in velocity. Id. at 60–61 (citing Ex. 1008, 491; Ex. 1003 ¶ 275). Petitioner explains that “Urmson’s speed IPR2019-00250 Patent 7,979,174 B2 35 planner is further programmed to calculate a desired speed based on a constraint that limits acceleration/deceleration.” Id. at 60 (citing Ex. 1008, 491; Ex. 1003 ¶¶ 275–276). And lastly, Petitioner alleges that “Urmson also discloses that its vehicles competing in the DARPA Challenge were constrained by a route data definition file (RDDF) containing ‘a series of waypoints marking the corridor and speed limits within which the robots were required to travel.’” Id. (citing Ex. 1008, 469; Ex. 1003 ¶ 277). Therefore, Petitioner explains that the speed planner calculates speeds based on the speed limits, i.e., constraints, from the route data definition file. Id. Petitioner further argues that Fregene discloses “nominal constraints,” which also meet the additional limitation of claim 12. Pet. 41. Petitioner contends that “Fregene explains that the ‘nominal constraints are more restrictive than the actual operating limits of the vehicle,’ since maximum performance, such as maximum vehicle speed, is only used in emergency situations.” Id. at 41–42 (quoting Ex. 1006 ¶¶ 25, 38). Petitioner asserts that because “under normal operation of the vehicle, Fregene’s autonomous vehicle is limited to speeds that do not violate the disclosed nominal constraints (i.e., priority is given to the nominal constraints under normal operation of the vehicle),” Fregene suggests the additional limitation of claim 12. Id. at 42. Petitioner reasons that the skilled artisan would have had reason to modify the proposed combined system of Andersson and Urmson to include the prioritized constraints suggested by Fregene “because it would enable Andersson’s vehicle to operate using a more restrictive (i.e., less mechanically demanding and more efficient) set of constraints during routine, normal operation” and would “prevent[] dramatic accelerations and IPR2019-00250 Patent 7,979,174 B2 36 decelerations during normal vehicle operation, thus improving passenger/driver comfort in the autonomous vehicle.” Id. at 42–43 (citing Ex. 1003 ¶ 208). Petitioner explains that a “PHOSITA would have recognized that it would be simple to modify the Andersson-Urmson system to include such functionality, especially since Andersson’s system already uses individual driver models to limit or restrict the generated speed plan, and in turn, the vehicle’s operation” and further that such “modification would not require any structural changes to the Andersson-Urmson system.” Id. at 43 (citing Ex. 1003 ¶ 203). Relevant to claim 13, Petitioner alleges that Fregene discloses the additional limitation of “permitting violation of a lower priority constraint in order to avoid violating a higher priority constraint.” Id. at 44. Petitioner explains that Fregene assigns priorities to its constraints and that collision avoidance is a high priority constraint. Id. (citing Ex. 1006 ¶¶ 7, 9). For example, “Fregene’s autonomous vehicle has a ‘collision avoidance system’ that will ‘switch [the] vehicle into an emergency maneuver mode in the event that an object enters the buffer zone and stays there for more than a preset threshold time.’” Id. According to Petitioner, “in Fregene’s system, avoidance of an obstacle takes higher priority over normal operation of the vehicle,” thereby meeting the requirement of assigning priorities to one or more constraints. Id. (citing Ex. 1006 ¶¶ 7, 9, 38; Ex. 1003 ¶¶ 198–199). Petitioner further explains that Fregene allows violation of lower priority constraints, such as speed, acceleration, and deceleration, in favor of collision avoidance, a higher priority constraint, during emergency operation. Id. at 44–45 (citing Ex. 1006, 25, 47). Petitioner asserts that one skilled in the art would have had reason to include Fregene’s functionality in IPR2019-00250 Patent 7,979,174 B2 37 the proposed combination of Andersson and Urmson in order “to improve its autonomous vehicle’s ability to avoid collisions” which, in turn, “improv[es] passenger/driver comfort.” Id. at 43, 46 (citing Ex. 1003 ¶¶ 201–203). Patent Owner disagrees and argues that Fregene fails to disclose either assigning priorities to constraints or allowing a lower priority constraint to be violated in favor of a higher priority constraint, and further argues a lack of a motivation to combine Fregene with the system of Andersson and Urmson. See PO Resp. 33–35. After reviewing the parties’ briefing and evidence of record, we are persuaded by Petitioner’s arguments and supporting evidence for claims 12 and 13, as summarized above. Accordingly, we determine that Petitioner has established by a preponderance of the evidence that each of the limitations of claims 12 and 13, except those noted as being challenged by Patent Owner, are present in the combination of Andersson, Urmson, and Fregene. See NuVasive, 841 F.3d at 974. We address Patent Owner’s specific arguments below. In addition to the reasons advanced for claim 9, Patent Owner argues that “none of Andersson, Urmson or Fregene disclose assigning priorities to constraints and violating a lower priority constraint in order to avoid violating a higher priority constraint as recited in claims 12 and 13.” PO Resp. 33–36.18 Specifically, Patent Owner alleges that “Fregene explicitly 18 Patent Owner references both claim 12 and 13 in arguing that Fregene does not teach violating a lower priority constraint in order to avoid violating a higher priority constraint. See generally PO Resp. 33–35. As noted above, claim 12 requires only “calculating the desired speed based on one or more prioritized constraints . . . [and] that the autonomous vehicle does not violate the one or more prioritized constraints” (Ex. 1001, 10:38– 41), whereas claim 13 concerns violating lower priority constraints in favor of higher priority constraints. Patent Owner acknowledges that “Fregene is, IPR2019-00250 Patent 7,979,174 B2 38 does not violate constraints at all: ‘this set associated with the emergency constraints of the vehicle is positively invariant,’ which ‘is entirely different from the concept embodied in the ‘174 patent and claims 12 and 13—the concept that some constraints are better to violate than others.”’ Id. at 34 (citing Ex. 1006 ¶ 25; Ex. 2001 ¶¶ 61, 140–141). Patent Owner explains that the nominal and emergency constraints taught by Fregene are two sets of operations that are alternatives to one another and “[w]ithin the set of constraints applied, Fregene makes clear that the constraints are not violated and does not disclose or contemplate a situation where its system confronts two constraints at the same time.” Sur-reply 12. Further, Patent Owner alleges that Fregene does not prioritize constraints for speed calculation. Id. We find persuasive Petitioner’s evidence and argument that Fregene suggests both assigning priorities to different constraints and that lower priority constraints may be violated in favor of higher priority constraints. See Ex. 1006 ¶¶ 7, 9, 25, 38, 47, 67, 72; Ex. 1003 ¶¶ 198–205, 255; Ex. 1025, 418:19–419:9, 420:10–13, 421:1–5; Pet. 61–65, 73; Pet. Reply 21–24. Fregene explains that two types of constraints are employed: The first constraint type includes the constraints under normal operation of the vehicle, herein called the nominal constraints. The second constraint type includes the constraints under an emergency operation of the vehicle, herein called the emergency constraints. The nominal constraints are more restrictive than the actual operating limits of the vehicle, since maximum performance is used for a vehicle only in emergency situations. however, consistent with the generally understood concept that constraints are generally not violated.” PO Resp. 34 (citing Ex. 2001 ¶¶ 140–141). IPR2019-00250 Patent 7,979,174 B2 39 Ex. 1006 ¶ 38. And Fregene explains that while the set associated with either of the normal constraints and the emergency constraints are positively invariant, “the intersection of two positively invariant sets is not positively invariant.” Id. ¶ 37. When the collision avoidance system of the normal mode fails, “[t]he vehicles are then operable to switch to an emergency maneuver mode and implement an emergency maneuver procedure.” Id. ¶ 53. Then when the emergency has passed, “[t]he on-board processor determines what instruction the vehicle requires to switch out of the emergency maneuver mode and back to normal operation mode.” Id. ¶ 64. Fregene adjusts vehicle speed depending on which constraints are applied. Id. ¶¶ 25, 38, 43, 52, 58, 71–74. Whether viewed as a set of constraints for different operational modes or as single set of constraints, the nominal constraints yield to emergency constraints when the vehicle transitions between normal and emergency mode. Therefore, we determine Petitioner has demonstrated by a preponderance of the evidence that Fregene suggests “calculating the desired speed based on one or more prioritized constraints such that the autonomous vehicle does not violate one or more prioritized constraints” and calculating the desired speed to “permit[] violation of a lower priority constraint in order to avoid violating a higher priority constraint,” as required by claims 12 and 13 respectively. Patent Owner further argues that “[a] POSA would also not be motivated to combine Fregene with Andersson and Urmson.” PO Resp. 34 (citing Ex. 2001 ¶¶ 146–152, 189). First, Patent Owner asserts that Andersson teaches away from a prioritized constraint system because it does not contemplate obstacle avoidance or discuss constraint priorities. Id. IPR2019-00250 Patent 7,979,174 B2 40 We disagree with Patent Owner’s argument that Andersson teaches away from a prioritized constraint system. As we explained above, Section II.D.4.d, silence does not imply teaching away. Allergan, 754 F.3d at 964. That Andersson is purportedly silent regarding prioritized constraint systems or obstacle avoidance does not constitute a teaching away because such disclosure, or lack thereof, does not criticize, discredit, or otherwise discourage the solution claimed in the ’174 patent. Second, Patent Owner argues that given the difficulties associated with the complexity of autonomous vehicle systems and Urmson’s teaching away from added complexities, a person of skill in the art would not have had reason to modify Urmson to include the priority constraint system of Fregene. PO Resp. 34–35 (citing Ex. 2001 ¶¶ 146–152, 189). For the reasons discussed above in Section II.D.4.d, we do not agree with Patent Owner’s arguments in this regard. And third, Patent Owner contends that “Fregene is also directed to obstacle avoidance systems for vehicles which is not in the same technological field as Andersson’s cruise control technology or Urmson’s autonomous vehicle technology.” Id. at 35 (citing Ex. 2001 ¶¶ 146–152, 189). Patent Owner also argues that “Fregene’s system is based in a controller external to the vehicle and a POSA would simply not combine such an external controller with the on-board vehicle systems of Andersson and Urmson.” Id. (citing Ex. 2001 ¶¶ 146–152, 189). For the reasons discussed above in Section II.D.4.d, we disagree with Patent Owner’s arguments relating to the differences between the technologies described in Andersson, Urmson, and Fregene. Further, we also disagree with Patent Owner’s position that Fregene utilizes an external IPR2019-00250 Patent 7,979,174 B2 41 processor for certain tasks such that a person skilled in the art would not seek out the teachings of Fregene. We note that Fregene uses a combination of an external processor, an onboard processor, and vehicle controllers for implementing its collision avoidance system. Ex. 1006 ¶ 8. Furthermore, Petitioner does not propose to substitute Fregene’s external processor into the combination of Andersson and Urmson; rather, Petitioner relies on Fregene to suggest a system where “one or more constraints is assigned a priority, wherein the speed planner is configured to allow violation of a lower priority constraint in order to avoid violation of a higher priority constraint.” Pet. 61. It is well established that a determination of obviousness based on teachings from multiple references does not require an actual, physical substitution of elements. In re Mouttet, 686 F.3d 1322, 1332 (Fed. Cir. 2012). Therefore, we determine that Petitioner has demonstrated by a preponderance of the evidence that the limitations of claims 12 and 13 would have been suggested by the combined disclosure of Andersson, Urmson, and Fregene, and that a person of ordinary skill in the art would have had reason to make the proposed combination with a reasonable expectation of success. F. Obviousness of Claims 14–17 in view of Urmson, Augenbraun, and Fregene Petitioner contends the subject matter of claims 14–17 would have been obvious over the combined disclosures of Urmson, Augenbraun, and Fregene. Pet. 55, 47. IPR2019-00250 Patent 7,979,174 B2 42 1. Augenbraun Augenbraun “relate[s] generally to the field of robotic devices, and more particularly to a multi-function robotic device having utility in various applications.” Ex. 1005 ¶ 3. “[M]ulti-function robotic device (‘robot’) 100 may generally be characterized by an ability to perform tasks autonomously or automatically with little or no intervention on the part of a user or operator.” Id. ¶ 49. Augenbraun describes robots having a body with a drive mechanism and a functional cartridge, coupled to the body, to operate “in accordance with characteristics of the functional cartridge.” Id. ¶¶ 10– 12; see also id. ¶¶ 13–14 (same). Embodiments exemplified in Augenbraun include functional cartridges “to perform a mopping function, a vacuuming function, or a sweeping function.” Id. ¶ 12. “The body may further comprise an electronics module to provide instructions to the drive mechanism to position the robot in an operating environment, and a sensing apparatus to provide data to the electronics module.” Id. ¶ 10. “The data may be related to a position of the robot relative to an object in the operating environment.” Id. The electronics module may include any suitable electronics “to provide necessary or desired control functions for robot 100.” Id. ¶ 57. Suitable electronics include “a microprocessor, an application specific integrated circuit (ASIC), a programmable logic controller (PLC), an arrangement of field programmable gate arrays (FPGA), or any other type of processor or addressable memory capable of providing instructions to influence the behavior of robot 100 or its components.” Id. The sensor array and a drive mechanism operate under the control of the electronics module, which “enable[s] robot 100 to navigate about an operating environment with precision, while avoiding obstacles.” Id. ¶ 76. The IPR2019-00250 Patent 7,979,174 B2 43 electronics module also signals the drive mechanism to permit “slowing or speeding one or both of motors 241 and 243 as necessary for desired for obstacle avoidance.” Id. 2. Analysis of Claim 14 Petitioner contends that independent claim 14 would have been obvious over the combination of Urmson, Augenbraun, and Fregene. Pet. 55. Claim 14 is directed to a programmable product including “program instructions embodied on a processor-readable medium for execution by a programmable processor” where the instructions cause the processor to (1) calculate a desired speed of a vehicle based on data that affects movement of the vehicle and prioritized conditions, and (2) output the calculated desired speed to a control system to calculate speed commands, which are used by actuators to adjust the speed of the vehicle. Ex. 1001, 10:46–58. Relevant to claim 14, Petitioner contends that Urmson describes a program product that includes program instructions on a programmable processor and that Augenbraun describes programmable processors, such as FPGAs. Pet. 55 (citing Ex. 1003 ¶ 264; Ex. 1005 ¶¶ 60, 76). Petitioner reasons that a person skilled in the art would have been motivated to modify Urmson’s system with Augenbraun’s processor “(i) to provide the specific hardware details for implementing a speed control module that are missing from Urmson, and (ii) to provide [a] more adaptable processor that can be programmed in the field after manufacture.” Id. at 63 (Ex. 1003 ¶ 280). Petitioner also explains that “Urmson discloses navigation software that includes a ‘speed planner’ … for calculating a desired speed for a vehicle.” Id. at 56 (citing Ex. 1008, Fig. 14). And, in particular, Petitioner argues that IPR2019-00250 Patent 7,979,174 B2 44 Urmson’s navigation software, utilizing RADAR and LIDAR as sensors to detect obstacles, includes a geometric planner to adjust the path as well as speed of the vehicle based on data received by the sensors, as required by claim 14. Id. at 57–58 (citing Ex. 1008, 479, 483, 486, 488, 491; Ex. 1003 ¶¶ 268, 270). Petitioner also alleges that Urmson’s navigation system is programmed to prioritize certain constraints. Id. at 58. By way of example, Petitioner directs our attention to Urmson’s disclosure that states the “speed planner ‘is responsible for ensuring driving speeds are safe’ and ‘slows [the autonomous vehicle] as necessary to swerve around an approaching obstacle.’” Id. at 59 (quoting Ex. 1008, 490–91; citing Ex. 1003 ¶ 274). “Therefore, Urmson’s speed planner is programmed to calculate a desired speed (i.e., a reduced speed) based on the high priority constraint of collision avoidance.” Id.; see also Ex. 1008, 491 (describing the speed planner algorithm); Ex. 1003 ¶¶ 275–276. Petitioner further contends that if Urmson is not understood to disclose prioritizing based on constraints, “such a deficiency is immaterial as a PHOSITA would know that the constraints in Urmson are ‘prioritized constraints.’” Pet. 61. Petitioner reasons that because Urmson desires safe driving speeds and collision avoidance, Urmson uses prioritized constraints that favor collision avoidance as the higher priority constraint. Id. (citing Ex. 1003 ¶ 278). Petitioner also argues that to the extent these “constraints must have ordered priorities,” a person of ordinary skill would have looked to Fregene, which “discloses [that] ‘calculating the desired speed includes calculating the desired speed based on one or more prioritized constraints such that the autonomous vehicle does not violate one or more prioritized IPR2019-00250 Patent 7,979,174 B2 45 constraints.’” Id. at 62 (referring to Petition Section IX(A); citing Ex. 1006 ¶¶ 38, 47 (describing nominal and emergency constraints); Ex. 1003 ¶ 279). Petitioner reasons that a person of ordinary skill in the art would have modified the Urmson-Augenbraun combination with Fregene’s prioritized constraints in order to “prevent dramatic accelerations and decelerations when imminent collisions (e.g., obstacles) are not present, thus improving a passenger’s comfort in the autonomous vehicle.” Id. at 63 (citing Ex. 1003 ¶ 281). Lastly, Petitioner asserts that Urmson’s speed controller receives the desired speed from the speed planner (described above) and calculates speed commands that enables actuators that throttle or break to adjust speed of the autonomous vehicle. Id. at 64–66 (citing Ex. 1008, 474–475, Fig. 9; Ex. 1003 ¶¶ 285–86). Patent Owner disagrees and contends that none of the references disclose “calculate a desired speed” or “prioritized constraints.” See PO Resp. 35. Patent Owner additionally argues that no motivation exists to modify the teachings of Urmson, Augenbraun, and Fregene. After reviewing the parties’ briefing and evidence of record, we are persuaded by Petitioner’s arguments and supporting evidence for the other limitations of claim 14 not challenged by Patent Owner as summarized above. Accordingly we determine that Petitioner has established by a preponderance of the evidence that each of these limitations of claim 14 are present in the combination of Urmson, Fregene, and Augenbraun. See NuVasive, 841 F.3d at 974. We address Patent Owner’s specific arguments below. IPR2019-00250 Patent 7,979,174 B2 46 Patent Owner contends that claim 14, and “specifically element 14[b], [which] recites ‘calculate a desired speed’ and ‘prioritized constraints,’ which are substantially similar to elements 9[b] and claim 12,” are not taught or suggested by Urmson or Fregene as alleged by Petitioner. PO Resp. 35; Sur-reply 17–18. For the reasons discussed above in addressing claims 9 and 12, we determine that Petitioner has demonstrated by a preponderance of the evidence that the subject matter of claim 14—that is, “calculate a desired speed” and “prioritized constraints”—is taught or suggested by the combination of Urmson and Fregene. Patent Owner further argues that a person of ordinary skill in the art “would not be motivated to modify Urmson or combine it as Petitioner suggests because of the complexity of autonomous vehicle systems such as Urmson, Urmson’s focus on simplicity (prioritized constraints are not a simple modification), and because the references are not in the same technological area.” PO Resp. 36. According to Patent Owner, because ASICs and FPGAs “require greater design and programming time, effort and cost,” a person of ordinary skill in the art would not seek out Augenbraun’s teachings. Id. We disagree with Patent Owner’s arguments that the skilled artisan would not have had reason to modify Urmson with Augenbraun. Petitioner relies on Augenbraun to teach a programmable processor to implement the navigation software of Urmson. Pet. 55. Urmson does not specify the details of how its navigation software is implemented. See generally Ex. 1008. Urmson explains that “[o]nboard navigation software combines incoming sensor data with a preplanned route to generate a new safe and traversable route.” Id. at 479. Urmson further explains that the “navigation IPR2019-00250 Patent 7,979,174 B2 47 software-architecture was designed with the infrastructure to support high- speed navigation.” Id. Augenbraun’s electronics module functions similarly to Urmson’s speed planner. Ex. 1005 ¶¶ 60, 76; Ex. 1003 ¶ 264. Augenbraun explains that electronics module 220 may include or be implemented as a microprocessor, an application specific integrated circuit (ASIC), a programmable logic controller (PLC), an arrangement of field programmable gate arrays (FPGA), or any other type of processor or addressable memory capable of providing instructions to influence the behavior of robot 100 or its components. Ex. 1005 ¶ 57. Thus, Petitioner uses Augenbraun to illustrate known programmable processors capable of implementing software such as the navigation software of Urmson. Pet. 55. Furthermore, as Dr. Reinholtz testifies, ASICs or FPGAs have certain advantages, i.e., low cost and low power consumption that must be weighed against disadvantages including difficulty in programming time and design. Ex. 2001 ¶¶ 125–126. As our reviewing court has recognized, “[a] given course of action often has simultaneous advantages and disadvantages, and this does not necessarily obviate motivation to combine.” See, e.g., Medichem, S.A. v. Rolabo, S.L., 437 F.3d 1157, 1165 (Fed. Cir. 2006); Winner Int'l Royalty, 202 F.3d at 1349 n.8. Patent Owner also argues that a person skilled in the art would not have had reason to combine Augenbraun with Urmson because “Augenbraun is directed to an entirely different technological area in that it involves small, differentially steered floor cleaning robots” where “Urmson, by contrast, involves full-size front-wheel-steered automobiles.” PO Resp. 37 (citing Ex. 2001 ¶¶ 127–128). We disagree with Patent Owner’s IPR2019-00250 Patent 7,979,174 B2 48 argument in this regard. Augenbraun embraces certain similarities that are relevant to the combination with Urmson. Specifically, Augenbraun describes an electronics module, similar to Urmson’s speed planner, which provides instructions to the drive mechanism to affect vehicle movement. See Ex. 1005 ¶¶ 10, 13; Ex. 1008, 479–480, Fig. 14. Furthermore, like Urmson, Augenbraun includes sensors to sense aspects of the operating environment, including obstacles in the vehicle path, which communicate with the drive mechanism to adjust progress of the vehicle. Ex. 1005 ¶¶ 60, 66–67, 76 (“[E]lectronics module 220 may provide appropriate signals to drive mechanism 240, slowing or speeding one or both . . . motors . . . as necessary or desired for obstacle avoidance . . . .”); Ex. 1008, 469 (“[T]he robots use[] onboard sensors to adjust a preplanned route, to avoid obstacles . . . .”), 476–479. Given these apparent similarities between Urmson and Augenbraun, a person of ordinary skill in the art would have had reason to look to references describing hardware and software implementations for autonomous vehicle control systems to implement program instructions. “[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.” KSR, 550 U.S. at 417. Accordingly, we determine that Petitioner has demonstrated by a preponderance of the evidence that the limitations of claim 14 would have been suggested by the disclosures of Urmson, Augenbraun, and Fregene, and that a person of ordinary skill in the art would have had reason to make the proposed combination with a reasonable expectation of success. IPR2019-00250 Patent 7,979,174 B2 49 3. Analysis of Claim 15 Claim 15 depends from claim 14 and requires that the data received regarding conditions which affect movement of the autonomous vehicle includes data regarding detected obstacles, wherein the program instructions are further operable to cause the programmable processor to calculate the desired speed such that the desired speed is a maximum safe speed when the received data indicates that no obstacles are detected and a minimum safe speed when the received data indicates that obstacles are detected. Ex. 1001, 10:59–67. Petitioner asserts Urmson ensures safe driving speeds and adjusts to slow the vehicle and swerve around obstacles and therefore, Urmson “teaches calculating and using a ‘maximum safe vehicle speed’ as the desired speed” and then slows the vehicle when an object is detected, i.e., the minimum safe speed, consistent with claim 15. Pet. 66–67 (quoting Ex. 1008, 490–491; citing Ex. 1003 ¶¶ 292–293). In addition to the arguments advanced for claim 14, Patent Owner argues that Urmson does not disclose a “minimum safe speed” as claimed. PO Resp. 37–38. Patent Owner urges that “Urmson does not disclose or render obvious speed calculation ‘based on the data received regarding conditions which affect movement of the autonomous vehicle’ as recited in element 14[d].” Id. at 38. We disagree with Patent Owner’s argument. Patent Owner offers little in the way of argument and does not specifically address Petitioner’s identified teaching regarding “minimum safe speed” or the contention that Urmson “affect movement of the autonomous vehicle,” by slowing the vehicle when objects are detected, which we find persuasive. See Pet. 56– 58. As explained above, the speed planner of Urmson ensures safe driving speeds and adjusts to slow the vehicle and swerve around obstacles. Ex. IPR2019-00250 Patent 7,979,174 B2 50 1008, 490–491; Ex. 1003 ¶¶ 144, 146, 148–49. In particular, Urmson explains that [w]hile the vehicle is driving, the algorithm runs on the output of the planner (a small subset of the overall path) allowing the vehicle to slow for unexpected obstacles. In practice, this algorithm performs well enough to drive safely on desert trails. As the vehicle approaches turns, it smoothly decelerates to a safe speed. As the vehicle approaches obstacles, it slows as necessary to swerve around them. Ex. 1008, 490–491. Both slowing the vehicle and swerving around obstacles “affect movement of the autonomous vehicle.” Therefore, we are persuaded by Petitioner’s arguments and supporting evidence regarding the additional limitations of claim 15, as summarized above, and find that Petitioner has shown, by a preponderance of the evidence, that Urmson suggests a “minimum safe speed” and that sensor data “affect[s] movement of the vehicle.” Accordingly, we determine that Petitioner has established by a preponderance of the evidence that the subject matter of claim 15 is suggested by the combination of Urmson, Augenbraun, and Fregene, and that the skilled artisan would have had reason to make the suggested proposed combination with a reasonable expectation of success. See Ex. 1008, 479, 483, 486, 488, 490–491, Fig. 14; Ex. 1003 ¶¶ 268, 270, 274–276, 278. 4. Analysis of Claims 16 and 17 Claims 16 and 17 depend from claim 14 and additionally require that the program instructions are further operable “to cause the programmable processor to calculate a desired path for each of a plurality of points along a planned path” and “to cause the programmable processor to calculate the desired speed by allowing violation of a lower priority constraint in order to IPR2019-00250 Patent 7,979,174 B2 51 avoid violation of a higher priority constraint”—similar to claims 10 and 13 discussed above. Ex. 1001, 11:1–17:3. Patent Owner does not substantively address Petitioner’s arguments with respect to claims 16 and 17 in its Patent Owner Response. See generally PO Resp. 38. Instead, Patent Owner argues that claims 16 and 17 are “patentable for the same reasons set forth above with regard to claim 14.” Id. With respect to dependent claims 16 and 17, Petitioner identifies where the combination of Urmson, Augenbraun, and Fregene suggests each limitation of these claims. Pet. 67–70. As to claim 16, Petitioner establishes that Urmson describes a geometric planner that calculates a planned path and the speed planner determines the maximum safe speed for points on the planned path, as required by claims 16. Id. at 67 (incorporating arguments in Section VIII(B) of the Petition). Regarding claim 17, Petitioner demonstrates that Fregene describes a prioritized constraint system, like that of claim 13 (discussed in Petition Section IX(B)), and that claim 17 additionally includes program instructions to cause a programmable processor to execute the instructions described by Augenbraun. Id. at 68. Petitioner also reasons, for claim 17, that “a PHOSITA would have been motivated to modify the Urmson-Augenbraun system to include functionality provided by Fregene’s system to improve its autonomous vehicle’s ability to avoid collisions, which is paramount in autonomous vehicle design, by enabling the autonomous vehicle to exceed normal constraints on vehicle performance in the event of an imminent collision.” Id. at 68–69 (citing Ex. 1003 ¶¶ 300–301). In addition, Petitioner reasons that a person skilled in the art would have included the lower and higher IPR2019-00250 Patent 7,979,174 B2 52 priority constraint system of Fregene “reserve maximum vehicle performance for emergency situations.” Id. at 69 (citing Ex. 1008, 491; Ex. 1003 ¶ 302). Petitioner explains that a modification of Urmson and Augenbraun to include the constraint system of Fregene would have been a simple modification because “Urmson’s speed planner already considers certain constraints” and “would merely require reprogramming the Urmson- Augenbraun system.” Id. at 69–70 (citing Ex. 1003 ¶ 303). For the reasons discussed above and in Sections II.D.5 and II.E., we find that Petitioner has demonstrated by a preponderance of the evidence that each of the additional limitations recited in claims 16 and 17 are present in or suggested by Urmson and Fregene, respectively. Accordingly, we determine that Petitioner has established by a preponderance of the evidence that the subject matter of claims 16 and 17 is suggested by the combination of Urmson, Anderson, and Fregene and that the skilled artisan would have had reason to make the proposed combination with a reasonable expectation of success. G. Obviousness of Claims 15 in view of Urmson, Augenbraun, Fregene, and Mandow Petitioner contends the subject matter of claim 15 would have been obvious over the combined disclosures of Urmson, Augenbraun, Fregene, and Mandow. Pet. 70. Petitioner includes Mandow “[i]n the unlikely event that ‘minimum safe speed’ is interpreted to be a slowest possible speed that is safe.” Id. Neither party contends “minimum safe speed” is the “slowest possible speed that is safe.” Rather, Petitioner does not offer a construction of this term (see generally Pet.), and Patent Owner contends the term means the “desired speed when obstacles are present” (PO Resp. 14, 39). IPR2019-00250 Patent 7,979,174 B2 53 Accordingly, we need not address Petitioner’s alternative position, which is based on a construction not advanced by either party. III. CONCLUSION19 We conclude that Petitioner has satisfied its burden of demonstrating, by a preponderance of the evidence, that the subject matter of claims 9–17 of the ’174 patent is unpatentable. In summary: 19 Should Patent Owner wish to pursue amendment of the challenged claims in a reissue or reexamination proceeding subsequent to the issuance of this decision, we draw Patent Owner’s attention to the April 2019 Notice Regarding Options for Amendments by Patent Owner Through Reissue or Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg. 16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application or a request for reexamination of the challenged patent, we remind Patent Owner of its continuing obligation to notify the Board of any such related matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2). 20 As explained above, we do not reach Petitioner’s alternative grounds including Ahmed-Zaid, because they are premised on a claim construction position not advanced by either party. Claims 35 U.S.C. § Reference(s)/Basis Claims Shown Unpatentable Claims Not shown Unpatentable 9–11 103(a) Andersson, Urmson 9–11 12, 13 103(a) Andersson, Urmson, Fregene 12, 13 9–11 103(a) Andersson, Urmson, Ahmed- Zaid20 12, 13 103(a) Andersson, Urmson, Ahmed- Zaid, Fregene IPR2019-00250 Patent 7,979,174 B2 54 IV. ORDER Accordingly, it is: ORDERED that Petitioner has established, by a preponderance of the evidence, that claims 9–17 of the ’174 patent are unpatentable; and FURTHER ORDERED that because this is a final written decision, parties to the proceeding seeking judicial review must comply with the notice and service requirements of 37 C.F.R. § 90.2. 21 As explained above, we do not reach Petitioner’s alternative ground including Mandow, because it is premised on a claim construction position not advanced by either party. 14–17 103(a) Urmson, Augenbraun, Fregene 14–17 15 103(a) Urmson, Augenbraun, Fregene, Mandow21 Overall Outcome 9–17 IPR2019-00250 Patent 7,979,174 B2 55 PETITIONER: Qingyu Yin Joshua Goldberg qingyu.yin@finnegan.com joshua.goldberg@finnegan.com PATENT OWNER: John Abramic Katherine Johnson Timothy C. Bickham Matthew Bathon Harold H. Fox Katherine D. Cappaert Steptoe & Johnson LLP jabramic@steptoe.com kjohnson@steptoe.com tbickham@steptoe.com mbathon@steptoe.com hfox@steptoe.com kcappaert@steptoe.com Copy with citationCopy as parenthetical citation