DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Respond to Amendment
This action is in response to the application filed on ---- 4/10/2026 for application 18/196,264. Claim 1 – 20 are pending and have been examined.
Claim 1, 11 and 19 are amended.
Respond to Argument
Applicant's argument filed on 4/10/2026 has been fully considered but they are not persuasive.
Regarding Claim rejection under 35 U.S.C. 103 section, applicant argued In page 11 – 12 that “combination of Shi and Huang fails to teach or suggest the newly claimed limitation: ’correct an output of at least one of an accelerometer or a gyroscope of the measurement unit based on the second magnetism data’”, “Huang is entirely silent on utilizing the neural-network-corrected magnetism data to actively correct the distinct outputs of an accelerometer or gyroscope. Similarly … Shi does not teach or suggest the specific claimed architecture where a trained computational network corrects magnetometer bias and then uses that specific computationally-derived data to correct the outputs of the accelerometer or gyroscope”, “A person of ordinary skill in the art combining Shi and Huang would, at best, arrive at an inertial measurement unit with a better-calibrated magnetometer. They would not arrive at the claimed system,”. Examiner notes that one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). Shi teaches: “orientation estimation … result will drift away with time because of gyroscope measurement errors… the magnetometer output is used to correct the yaw estimation” (Shi, Sec. I). Hung teaches “In order to accurately measure the three-component and its modulus value in earth's magnetic field, need to carry out the error school of strapdown three axis magnetometer Just” (Hung translation page 5), “Using training the neural network weight successfully obtained to constitute error corrector, error correction is carried out to the instrument error of three axis magnetometer” (Hung translation page 1). i.e., Hung points out that the magnetometer readings need correction and the solution of Hung is using neural network to perform the correction. The combination of Shi and Hung renders obviousness to corrected magnetism data before correcting the gyro sensor data. For further detail, see claim rejection under 35 U.S.C. 103 section.
The remaining arguments are essentially the same as those addressed above and/or below and are unpersuasive for at least the same reasons. Therefore, Examiner is unpersuaded and maintains the corresponding rejections.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1, 6, 8 – 9, 11 and 18 – 20 are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al., (hereinafter Shi), “An Improved Yaw Estimation Algorithm for Land Vehicles Using MARG Sensors” in view of Huang et al., (hereinafter Huang), CN107290801,
Regarding Claim 1, Shi discloses: A system comprising: a memory configured to store computer executable instructions; and one or more processors configured to execute the instructions to: obtain a set of measurement unit attributes and location information associated with a measurement unit, the set of measurement unit attributes comprising a first magnetism data for the measurement unit (Shi, sec. 4.1, “Global Position System (GPS) … magnetic sensor”), wherein the first magnetism data is generated based on a magnetometer of the measurement unit subject to a magnetic bias (page 3, “Magnetic disturbances include hard iron effects, soft iron effects (magnetic bias)”);
correct an output of at least one of an accelerometer or a gyroscope of the measurement unit based on the … magnetism data (Shi, sec 1, page 2, “the magnetometer output is used to correct the yaw estimation”; yaw is a gyroscope reading, Shi teaches using magnetometer reading/output to correct yaw estimate).
Shi does not explicitly disclose:
identify a plurality of reference magnetism sources in proximity of the measurement unit, based on the location information;
obtain reference magnetism data from the plurality of reference magnetism sources; and
generate second magnetism data for the measurement unit, based on the first magnetism data, the reference magnetism data, and a trained computational network,
wherein the trained computational network corrects an inaccuracy of the first magnetism data based on the reference magnetism data to generate the second magnetism data, and
wherein the inaccuracy is due to the magnetic bias of the magnetometer.
correct an output of at least one of an accelerometer or a gyroscope of the measurement unit based on the second magnetism data
Huang, in the same field of endeavor, explicitly teach:
identify a plurality of reference magnetism sources in proximity of the measurement unit, based on the location information (Huang, translation page 6, “chooses calibration region of the stable open area in one piece of earth's magnetic field … strapdown three Axis magnetometer is rotated rotating around three axis of carrier, obtains the geomagnetic field measuring data under different sensors posture”; at a certain location, measuring magnetism data of different postures (different sources). );
obtain reference magnetism data from the plurality of reference magnetism sources (refer to the mapping above, the magnetism data of different postures are collected. Examiner notes that 0046 of specification of instant application described the reference magnetism source as: “The term ‘reference magnetism source’ may refer to a reference source … that may monitor magnetism value or magnetic field values for a corresponding geographic region” for reference; Huang teaches the collection/monitoring of magnetism value for a corresponding geographic region by different postures for reference, and thus is analogous); and
generate second magnetism data for the measurement unit, based on the first magnetism data, the reference magnetism data and a trained computational network (Huang, eq. 1 – 2, 0011 – 0012, & translation page 6, “by neural network weight (trained computational network) … The inverse matrix Ω of strapdown three axis magnetometer calculation matrix is calculated by formula 1 … The equivalent zero bias of strapdown three axis magnetometer are calculated by formula 2”, “Obtain the earth magnetism (second magnetism data) after error correction Field estimated value (first magnetism data)”; i.e., the computational network is trained with reference magnetism data and is to generate a non-biased magnetism data (second magnetism data)),
wherein the trained computational network corrects an inaccuracy of the first magnetism data based on the reference magnetism data to generate the second magnetism data (refer to the mapping above, the network is to correct the bias (inaccuracy) of the field estimated value), and
wherein the inaccuracy is due to the magnetic bias of the magnetometer (Huang translation page 5, “the three axis magnetometer such as not perfect of manufacturing technology and mounting process are deposited In instrument errors such as three axis are nonopiate, between centers scale factor deviations and zero bias; The usual strapdown of three axis magnetometer is in carrier, therefore three There is installation alignment error between axis magnetometer and carrier; There are also the interference such as certain hard iron magnetic field and soft iron magnetic field for ambient environment Magnetic field. These instrument errors and interference magnetic field will affect strapdown three axis magnetometer to the measurement essence of earth's magnetic field three-component and its modulus value Degree.”).
correct an output of at least one of an accelerometer or a gyroscope of the measurement unit based on the second magnetism data (Hung translation page 5, “In order to accurately measure the three-component and its modulus value in earth's magnetic field, need to carry out the error school of strapdown three axis magnetometer Just”; i.e., Hung points out that the magnetometer readings have error and need correction. The combination of Shi and Hung renders obviousness to corrected magnetism data before correcting the gyro sensor data.)
Shi and Huang both teach application of magnetometer and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to further include the calibration techniques taught by Huang in the system of Shi to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification in order increase the accuracy of measuring data (Huang translation page 5 – 6).
Regarding Claim 6. Shi and Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach: the measurement unit is associated with at least one of: a vehicle, a compass, or an exploration system (Shi, page 1, “estimation for a land vehicle”).
Regarding Claim 8. Shi, Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach: the measurement unit is at least one of: a nine-axis inertial measurement unit, a six-axis inertial measurement unit, or a three-axis inertial measurement unit (Shi, page 12, “3-axis gyro module”; fig. 4 – 5, the sensor is to detect the yaw rate and thus 3-axis inertial measurement unit).
Regarding Claim 9. Shi, Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach: the set of measurement unit attributes comprise at least one of: the first magnetism data, speed data, orientation data, force data, acceleration information, bearing information, and angular rate information (refer to the mapping of Claim 8, yaw rate is the angular rate information ).
Regarding Claim 11 & 18, these are the corresponding method claim of Claim 1 & 6. These claims are rejected with the same reason.
Regarding Claim 19 – 20, these are the corresponding computer programmable product claim corresponding to Claim 1 – 2. Shi further teaches: a non-transitory computer readable
medium having stored thereon computer executable instructions, which when executed by one or more processors, cause the one or more processors to conduct operations (Shi, fig. 3b, the operation is carried out by a laptop). Claim 19 – 20 are rejected with the same reason.
Claim(s) 2 – 4, 12 – 14 are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al., (hereinafter Shi), “An Improved Yaw Estimation Algorithm for Land Vehicles Using MARG Sensors” in view of Huang et al., (hereinafter Huang), CN107290801, as applied to claim 1 above, and further in view of measure.ca, “Trinity F90: How to Calibrate The Magnetometer” with evidential reference of Rainier Lamers “Do Magnetometers Need to be Recalibrated from Time to Time?”.
Regarding Claim 2. Shi and Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach:
obtain historical magnetism data … generate the first magnetism data for a first geographical region of the measurement unit based on the location information and the trained computational network, the trained machine- learning based computational network being trained on the historical magnetism data (refer to the mapping in Claim 1 & translation page 6, the collected magnetism data of different postures is the historic magnetism data of the geographic location. The model is machine learning computational network trained on the collected magnetism data)
transmit the first magnetism data to the measurement unit, wherein the measurement unit generates the set of measurement unit attributes using the first magnetism data (refer to the mapping in Claim 1 and above, Shi teaches using magnetism data to correct/generate inertial sensor data; Huang teaches generate corrected magnetism data using field estimated value (first magnetism data). The combined teach renders obviousness of the limitation).
The combination does not explicitly teach:
obtain historical magnetism data for a plurality of geographical regions;
Measure.ca, in the same field of endeavor, explicitly teach:
obtain historical magnetism data for a plurality of geographical regions (Measure.ca, page 1, “The magnetometer needs to be re-calibrated when the actual flight location is more than 50km away from the location where the magnetometer was calibrated before”; Measure.ca teaches recalibrate magnetometer at different location, Huang teaches use training/retraining computational network as the calibration step. The combination renders obviousness of the claimed limitation. Examiner further notes that recalibration of magnetometer at different location is a known step in the field. To extend the evidence, examiner include additional reference: Rainier Lamers “Do Magnetometers Need to be Recalibrated from Time to Time?”);
Shi (in view of Huang) and Measure.ca both teach application of magnetometer and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to further include the frequent calibration taught by Measure.ca in the system of Shi (in view of Huang) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification in order to adapt the magnetometer components to the location (Rainier Lamers, page 2).
Regarding Claim 3. Shi and Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach: determine magnetism accuracy based on the reference magnetism data and the first magnetism data; and train the machine-learning based computational network based on the magnetism accuracy, the reference magnetism data, the set of measurement unit attributes and the first magnetism data (refer to the mapping in Claim 1 & Huang fig. 1, the reference data is used to train the model. The error (accuracy) between the model output and the collected reference is the feedback in the loop during training).
Shi and Huang combination does not explicitly teach:
re-train the trained machine-learning based computational network
Measure.ca, in the same field of endeavor, explicitly teach:
re-train the trained machine-learning based computational network (refer to the mapping in Claim 2, Measure.ca suggest retraining of the trained model. Thus the combination renders obviousness of the claimed limitation.)
The reason for combination is same as Claim 2.
Regarding Claim 4. Shi, Huang and Measure.ca combination renders obviousness of all the limitation in Claim 3. The combination further teach: generate the second magnetism data for the measurement unit, using the re-trained computational network; transmit the second magnetism data to the measurement unit (refer to the mapping of Claim 3. Retrained model generate second magnetism data); and cause to generate, by the measurement unit, updated set of measurement unit attributes (refer to the mapping of Claim 1 & Shi, fig. 2, & page 3 “improve the yaw estimation … the method implements measurement updates … to eliminate its impact on attitude estimation”; i.e., the magnetism data is used to improve/update the yaw estimation).
Regarding Claim 12 – 14 these are the corresponding method claim of Claim 2 – 4. These claims are rejected with the same reason.
Claim(s) 5 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al., (hereinafter Shi), “An Improved Yaw Estimation Algorithm for Land Vehicles Using MARG Sensors” in view of Huang et al., (hereinafter Huang), CN107290801, as applied to claim 1 above, and further in view of InterMagnet, “International Real-time, Magnetic Observatory Network”, and NCEI, “CrowdMag“.
Regarding Claim 5. Shi, and Huang combination renders obviousness of all the limitation in Claim 1. The combination does not explicitly teach: the plurality of reference magnetism sources comprises one or more reference magnetic stations and one or more crowd data sources.
InterMagnet, in the same field of endeavor, explicitly teach: the plurality of reference magnetism sources comprises one or more reference magnetic stations (InterMagnet, page 1, “you can find data and information from geomagnetic observatories (magnetism source) around the world … for measuring and recording equipment, in order to facilitate data exchanges … in close to real time”; InterMagnet is a shared database of magnetism measurement. The data is collected by each observatories);
Shi (in view of Huang) and InterMagnet both teach the use of magnetism data and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to utilize the data from observatories as suggested by InterMagnet in the system of Shi (in view of Huang) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification to acquire training data.
Shi, Huang and InterMagnet combination does not explicitly teach: one or more crowd data sources.
NCEI, in the same field of endeavor, explicitly teach:
one or more crowd data sources (NCEI, page 1, “crowdsourced data collection project that uses a mobile app to collect geomagnetic data from magnetometers that modern smartphone use … the data have the potential to provide a high resolution alternative … real-time information about changes in the magnetic field”).
Shi (in view of Huang and InterMagnet) and CrowdMag both teach the use of magnetism data and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to utilize the data from crowd source as suggested by CrowdMag in the system of Shi (in view of Huang and InterMagnet) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification to acquire up-to-date and accurate geomagnetic field data (CrowdMag, page 2).
Regarding Claim 17, Claim 17 is the corresponding method claim of Claim 5. Claim 17 is rejected with the same reason.
Claim(s) 7 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al., (hereinafter Shi), “An Improved Yaw Estimation Algorithm for Land Vehicles Using MARG Sensors” in view of Huang et al., (hereinafter Huang), CN107290801, as applied to claim 1 above, and further in view of Shashua, WO2016130719.
Regarding Claim 7. Shi and Huang combination renders obviousness of all the limitation in Claim 1. The combination does not explicitly teach: generate navigation instructions based on the second magnetism data; and update a map database based on the second magnetism data and the generated navigation instructions.
Shashua, in the same field of endeavor, explicitly teach:
generate navigation instructions based on the second magnetism data (Shashua, 0435, “In some embodiments, the trajectory may be reconstructed based on data from inertial sensors”; Shi teaches a vehicle system that the inertial reading is based on magnetism data. Shashua teaches generating trajectory (navigation instruction) based on inertial reading. The combination renders obviousness of the limitation); and update a map database based on the second magnetism data and the generated navigation instructions (Shashua, 0114 – 0116, “determine … existence in the environment of the vehicle of a navigational adjustment condition; cause the vehicle to adjust the navigational maneuver … transmitted to a road model management system for determining whether an update to the predetermined model representative of the road segment”; The environment sensing is based on the vehicle trajectory thus the updating of the map is also based on the generated navigation instructions and the magnetism data).
Shi (in view of Huang) and Shashua both teach the control of autonomous driving and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to include the mapping function of Shashua in the system of Shi (in view of Huang) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification to keep the road information up to date.
Regarding Claim 15, Claim 15 is the corresponding method claim of Claim 7. Claim 15 is rejected with the same reason.
Claim(s) 10 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al., (hereinafter Shi), “An Improved Yaw Estimation Algorithm for Land Vehicles Using MARG Sensors” in view of Huang et al., (hereinafter Huang), CN107290801”, as applied to claim 1 above, and further in view of Measure.ca, “Trinity F90: How to Calibrate The Magnetometer” and Goodfellow, “Generative Adversarial Nets”, with evidential reference of Rainier Lamers “Do Magnetometers Need to be Recalibrated from Time to Time?”.
Regarding Claim 10. Shi, Huang combination renders obviousness of all the limitation in Claim 1. The combination further teach: to train the computation network, the one or more processors are further configured to:
receive training data comprising historical magnetism data for a … geographical regions, the historical magnetism data comprising at least one of: one or more historical magnetometer readings, one or more crowd-sourced magnetometer readings, or one or more historical reference magnetism data (refer to the mapping in Claim 1 & Huang, the collected magnetic data of different posture is the historical magnetism data/reading.);
determine a plurality of features corresponding to magnetism for … geographical regions, using the training data (refer to the mapping above, the corrected magnetic data is the feature corresponding to magnetism for the location); and
The combination does not explicitly teach:
for a plurality of geographical regions
for each of the plurality of geographical regions
train the machine-learning based computational network to generate test magnetism value for one or more test geographical regions, using the plurality of features and the set of historical magnetism data
Measure.ca, in the same field of endeavor, explicitly teach:
for a plurality of geographical regions … for each of the plurality of geographical regions (Measure.ca, page 1, “The magnetometer needs to be re-calibrated when the actual flight location is more than 50km away from the location where the magnetometer was calibrated before”; Measure.ca teaches recalibrate magnetometer at different location, Huang teaches use training/retraining computational network as the calibration step. The combination renders obviousness of the claimed limitation. Examiner further notes that recalibration of magnetometer at different location is a known step in the field. To extend the evidence, examiner include additional reference: Rainier Lamers “Do Magnetometers Need to be Recalibrated from Time to Time?”);
Shi (in view of Huang) and Measure.ca both teach application of magnetometer and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to further include the frequent calibration taught by Measure.ca in the system of Shi (in view of Huang) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification in order to adapt the magnetometer components to the location (Rainier Lamers, page 2).
Shi, Huang and Measure.ca combination does not explicitly teach:
train the machine-learning based computational network to generate test magnetism value for one or more test geographical regions, using the plurality of features and the set of historical magnetism data.
Goodfellow, in the same field of endeavor, explicitly teach:
train the machine-learning based computational network to generate test magnetism value for one or more test geographical regions, using the plurality of features and the set of historical magnetism data (Goodfellow, sec. 5, table 1, “The reported numbers on MNIST are the mean loglikelihood of samples on test set”; i.e., the model is tested on the testing dataset after the training is complete to verify the performance of the trained model).
Shi (in view of Huang and Measure.ca) and Goodfellow both teach the training of generative adversarial network model and are analogous. It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention with a reasonable likelihood of success to further include the testing step of Goodfellow in the system of Shi (in view of Huang and Measure.ca) to achieve the claimed teaching. One of the ordinary skill in the art would have motivated to make this modification to verify the performance and give user confidence before the model is deployed.
Regarding Claim 16, Claim 16 is the corresponding method claim of Claim 10. Claim 16 is rejected with the same reason.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure: Wu, “Magnetometer and Gyroscope Calibration Method with Level Rotation” which teaches that “integrated calibration methods for gyroscopes and magnetometers have been widely developed in the literature” and propose a filtered calibration model for both Magnetometer and Gyroscope.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHIEN MING CHOU whose telephone number is (571)272-9354. The examiner can normally be reached Monday- Friday 9 am - 5 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, HITESH PATEL can be reached on 571-270-5442. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/SHIEN MING CHOU/Examiner, Art Unit 3667
/Hitesh Patel/Supervisory Patent Examiner, Art Unit 3667
6/24/26
Prosecution Insights