Harshavardhana Kikkeri et al.

Patent Trials and Appeals BoardAug 29, 2019

14146494 - (D) (P.T.A.B. Aug. 29, 2019)

UNITED STATES PATENT AND TRADEMARK OFFICE UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www.uspto.gov APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO. 14/146,494 01/02/2014 Harshavardhana Kikkeri 340256-US-NP 7153 69316 7590 08/29/2019 MICROSOFT CORPORATION ONE MICROSOFT WAY REDMOND, WA 98052 EXAMINER SMITH-STEWART, DEMETRA R ART UNIT PAPER NUMBER 3661 NOTIFICATION DATE DELIVERY MODE 08/29/2019 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): chriochs@microsoft.com usdocket@microsoft.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE ____________ BEFORE THE PATENT TRIAL AND APPEAL BOARD ____________ Ex parte HARSHAVARDHANA KIKKERI, STANLEY T. BIRCHFIELD, and MIHAI JALOBEANU ____________ Appeal 2017-010028 Application 14/146,4941 Technology Center 3600 ____________ Before EDWARD A. BROWN, JEREMY M. PLENZLER, and SEAN P. O’HANLON, Administrative Patent Judges. BROWN, Administrative Patent Judge. DECISION ON APPEAL STATEMENT OF THE CASE Appellant seeks review under 35 U.S.C. § 134(a) of the Examiner’s decision rejecting claims 1–21 under 35 U.S.C. § 102(a)(1) as anticipated by Gutmann.2 We have jurisdiction under 35 U.S.C. § 6(b). We REVERSE. 1 Microsoft Technology Licensing, LLC (“Appellant”) is the applicant and is identified as the real party in interest. 37 C.F.R. § 1.46; Appeal Br. 2. 2 US 2013/0138247 A1, published May 30, 2013. Appeal 2017-010028 Application 14/146,494 2 CLAIMED SUBJECT MATTER Claims 1, 10, and 18 are independent claims. Claim 1, reproduced below, illustrates the claimed subject matter. 1. A method for ground truth estimation for an autonomous navigation system (navigator), the method comprising: calibrating a ground truth estimation system for the navigator, by determining a calibration pose of a camera of the navigator as disposed in relation to each of a plurality of landmarks during a calibration period; directing the navigator to travel to a sequence of waypoints comprising a sequence of the landmarks during an evaluation period; and determining, using the calibrated ground truth estimation system, a ground truth estimation of the navigator based on an accuracy pose of the camera of the navigator as disposed in relation to the waypoints during the evaluation period, and the calibration poses disposed in relation to the plurality of landmarks. Appeal Br. 12 (Claims App.). ANALYSIS The Examiner finds that Gutmann discloses all limitations of independent claims 1, 10, and 18. Final Act. 4–6 (citing Gutmann ¶¶ 88, 124, 200, 204, 205, 209, 219, 222, 228–330, 399). Appellant notes that claim 1 recites “determining . . . a ground truth estimation . . . based on an accuracy pose of [a] camera . . . as disposed in relation to the . . . waypoints during . . . evaluation, and the calibration poses disposed in relation to the . . . landmarks [during calibration].” Appeal Br. 8 (hereinafter, “the ground truth estimation feature”). Claims 10 and 18 recite substantially similar language. Id. at 14–15, 17–18 (Claims App.). Appellant explains that claim 1 calls for calibrating an autonomous navigation system (navigator) during a Appeal 2017-010028 Application 14/146,494 3 calibration period using calibration poses of a camera, and evaluating the navigator during an evaluation period using accuracy poses of the camera. Id. at 8 (citing Spec. ¶¶ 51–55). Appellant explains that the ground truth estimation of the navigator is determined based on differences between the poses, that is, both calibration poses and accuracy poses. Id. (citing Spec. ¶ 55). Appellant contends that Gutmann does not disclose the claimed ground truth estimation feature. Appeal Br. 8. Appellant contends that Gutmann discloses using simultaneous localization and mapping (SLAM) to localize a mobile robot in an unknown environment and to re-localize the robot, based on continuous signals. Id. at 8–9 (citing Gutmann, Abstract, ¶¶ 58, 60, 79, 215). Appellant contends that Gutmann describes estimating a robot pose in an unmapped environment using SLAM (id. at 9 (citing Gutmann ¶¶ 379–383)), and estimating the pose of a camera (i.e., camera position and orientation) (id. (citing Gutmann ¶ 360)). Appellant explains that Gutmann discloses the robot projecting spots continuously onto a room ceiling, measuring a camera position of each spot to determine the camera position, and determining a rotation angle of the camera. Id. (citing Gutmann ¶¶ 215, 379–383). Appellant contends that estimating the pose of a camera or robot during localization and re-localization relates to determining the robot’s position, whereas the claims recite determining a ground truth estimation of the robot’s navigation system based on poses determined during calibration and evaluation periods. Id. Appellant also contends that Gutmann uses available ground truth information to determine the accuracy of individual SLAM implementations, and “ground truth information is not ground truth estimation,” but is merely Appeal 2017-010028 Application 14/146,494 4 a set a data. Appeal Br. 9; Reply Br. 2 (citing Gutmann ¶ 205). According to Appellant, Gutmann compares SLAM-derived poses to ground truth information describing actual poses to determine this accuracy. Reply Br. 2. Appellant maintains that the ground truth information does not include poses determined during calibration and evaluation periods, and thus, Gutmann does not disclose the claimed ground truth estimation feature. Id. The Examiner responds that Appellant’s Specification does not clearly define the term “estimation” in “ground truth estimation,” and “does not give any specific meaning as to what is the estimation of the robot’s navigation system.” Ans. 21. The Examiner finds that, according to paragraph 3 of the Specification, “the ground truth estimation includes calibrating a ground truth estimation system for the navigator, by determining a calibration pose of the navigator as disposed in relation to each of a plurality of landmarks during a calibration period.” Id. And, in response to Appellant’s contention that “Gutmann does not determine a ground truth estimation of the robot’s navigation system based on the poses of the camera” (Appeal Br. 9 (emphasis added)), the Examiner determines that “based on the direction measurement of spots on the ceiling of Gutmann, a calibration pose of the robot is determined.” Ans. 22 (emphasis added). Further, in response to Appellant’s contention that “ground truth estimation,” as claimed, is different from “ground truth information” described in Gutmann (Appeal Br. 9; Reply Br. 2), the Examiner states that it is unclear from the claims and Specification “what [is] the specific meaning of the estimation of ground truth that distinguishes Gutmann’s ground truth information.” Ans. 23. However, the Examiner then states, “[t]he specification defines ground truth estimation as Appeal 2017-010028 Application 14/146,494 5 determining a calibration pose of the navigator as disposed in relation to each of a plurality of landmarks during a calibration period.” Id. Even if Gutmann discloses determining a calibration pose of a robot, claim 1 requires “determining . . . a ground truth estimation . . . based on an accuracy pose of [a] camera . . . as disposed in relation to the . . . waypoints during . . . evaluation, and the calibration poses disposed in relation to the . . . landmarks [during calibration].” Appeal Br. 1 (Claims App.) (emphasis added). This language requires determining a ground truth estimation of the navigator based on both the calibration poses and accuracy poses of the camera. As claimed, the same navigator is used during the calibration period and the evaluation period, and the calibration poses and accuracy poses of the camera of the navigator are determined for common landmarks. The Examiner does not take into account that determining a ground truth estimation is also based on an accuracy pose of the camera during the evaluation period, as claimed. Additionally, the Examiner does not establish with evidence that the ground truth information described in Gutmann is determined based on an accuracy pose of a camera of a navigator as disposed in relation to waypoints during an evaluation period and calibration poses disposed in relation to a plurality of landmarks during a calibration period, as required by claim 1. We agree with Appellant that the Examiner has not established that the ground truth information described in Gutmann discloses the claimed ground truth estimation feature. Appellant states that “confirming a simultaneous localization and mapping (SLAM) implementation based on SLAM-derived poses and the actual poses is not the same as determining ground truth estimation,” as Appeal 2017-010028 Application 14/146,494 6 claimed. Reply Br. 2 (emphasis added). As discussed, “the actual poses” refers to the ground truth estimation information disclosed in Gutmann. We agree with Appellant that the Examiner has not established this equivalence by a preponderance of the evidence. Instead, the Examiner’s position that Gutmann anticipates the claims fails to take into account all that is required to determine a ground truth estimation in claims 1, 10, and 18. We do not sustain the rejection of claims 1–21 as anticipated by Gutmann. DECISION We reverse the rejection of claims 1–21. REVERSED