# US 11,897,486 B1 · Sensor consensus monitor > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,897,486 B1 - Document ID: US11897486B1 - Status: Granted - Kind: B1 - Application number: 17/536,940 - Filed: 2021-11-29 - Granted: 2024-02-13 - Earliest claimed: 2021-11-29 - Assignees: Zoox, Inc. - Applicants: Zoox, Inc. - Inventors: Derek Adams; Eric Andre Senant; Weihung Ko; Michael Carsten Bosse - Web page: https://patentgrep.com/patent/US11897486B1 ## CPC - B60W 50/0098: {Details of control systems ensuring comfort, safety or stability not otherwise provided for} - G01C 25/005: {initial alignment, calibration or starting-up of inertial devices} - B60W 60/001: {Planning or execution of driving tasks} - G01C 21/165: {combined with non-inertial navigation instruments} - G01C 21/1652: {with ranging devices, e.g. LIDAR or RADAR} - G01C 21/1656: {with passive imaging devices, e.g. cameras} - G01C 21/183: {Compensation of inertial measurements, e.g. for temperature effects} - G01C 21/30: Map- or contour-matching - B60W 2050/0055: {High-pass filters} - B60W 2050/0056: {Low-pass filters} - B60W 2050/0085: {Setting or resetting initial positions} - B60W: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) ## Abstract Techniques described are related to determining when a discrepancy between data of multiple sensors (e.g., IMUs) might be attributable to a sensor error, as opposed to operating conditions, such as sensor bias or noise. For example, the sensor data is passed through one or more filters (e.g., bandpass filter) that model the bias or noise, and the filtered data may then be compared for consistency. In some examples, consistency may be based on residuals or some other metric describing discrepancy among the filtered sensor data. ## Claims ### Claim 1 A method comprising: receiving first sensor data associated with a first sensor and second sensor data associated with a second sensor; determining third sensor data by filtering high-frequency data and low-frequency data from the first sensor data and the second sensor data; determining, based at least in part on the third sensor data, a discrepancy metric associated with the first sensor; determining, based at least in part on the discrepancy metric exceeding a threshold metric, an error associated with the first sensor; and using, based at least in part on the error associated with the first sensor, the second sensor data instead of the first sensor data to control an operation of a vehicle. ### Claim 2 (depends on 1) The method of claim 1, wherein the first sensor and the second sensor include gyroscopes, accelerometers, pressure sensors, magnetometers, inertial measurement units, or any combination thereof. ### Claim 3 (depends on 1) The method of claim 1 further comprising: receiving fourth sensor data associated with the first sensor, wherein the fourth sensor data is received prior to the first sensor data; and controlling, based on the fourth sensor data, a second operation of the vehicle before the operation, wherein the operation includes recalibrating the first sensor. ### Claim 4 (depends on 1) The method of claim 1, wherein: the filtering of high-frequency data includes using a low-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor noise; and the low-pass filter includes one or more of: an exponential filter, a finite impulse response filter, an infinite impulse response filter, or an exponential decay filter. ### Claim 5 (depends on 1) The method of claim 1, wherein: the filtering of low-frequency data includes using a high-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor bias; and the high-pass filter includes one or more of: an exponential filter, a finite impulse response filter, an infinite impulse response filter, or an exponential decay filter. ### Claim 6 (depends on 1) The method of claim 1, wherein the discrepancy metric includes a first residual that is associated with the first sensor and that is higher than a second residual associated with the second sensor. ### Claim 7 (depends on 1) The method of claim 1, wherein: the first sensor and the second sensor include a first inertial measurement unit (IMU) and a second IMU associated with respective positions relative to the vehicle; the method further comprises: determining, by transformation to a reference point associated with the vehicle, transformed sensor data based on the first sensor data and the second sensor data; and the discrepancy metric is based on the transformed sensor data. ### Claim 8 (depends on 1) The method of claim 1, wherein: the first sensor data and the second sensor data include time series data; and the method further comprises beginning to filter the high-frequency data at an earlier time in the time series data than a later time at which beginning to filter the low-frequency data. ### Claim 9 One or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform operations comprising: receive first sensor data associated with a first sensor and second sensor data associated with a second sensor; determine third sensor data by filtering high-frequency data and low-frequency data from the first sensor data and the second sensor data; determine, based at least in part on the third sensor data, a discrepancy metric associated with the first sensor; determine, based at least in part on the discrepancy metric exceeding a threshold metric, an error associated with the first sensor; and use, based at least in part on the error associated with the first sensor, the second sensor data instead of the first sensor data to control an operation of a vehicle. ### Claim 10 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein the first sensor and the second sensor include gyroscopes, accelerometers, pressure sensors, magnetometers, inertial measurement units, or any combination thereof. ### Claim 11 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein operations further comprise: receiving fourth sensor data associated with the first sensor, wherein the fourth sensor data is received prior to the first sensor data; and controlling, based at least in part on the fourth sensor data, a second operation of the vehicle before the operation; and changing, based at least in part on the error associated with the first sensor, from controlling the vehicle based on data generated by the first sensor to controlling the vehicle based on data generated by the second sensor. ### Claim 12 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein: the filtering of high-frequency data includes using a low-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor noise; the filtering of low-frequency data includes using a high-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor bias; and the low-pass filter and the high-pass filter comprise one or more of: an exponential filter, a finite impulse response filter, an infinite impulse response filter, an exponential decay filter, or a median filter. ### Claim 13 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein the discrepancy metric includes a first residual that is associated with the first sensor and that is higher than a second residual associated with the second sensor. ### Claim 14 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein: the first sensor and the second sensor include a first inertial measurement unit and a second IMU associated with respective positions relative to the vehicle; the operations further comprise determining, by transformation to a reference point associated with the vehicle, transformed sensor data based on the first sensor data and the second sensor data; and the discrepancy metric is based on the transformed sensor data. ### Claim 15 (depends on 9) The one or more non-transitory computer-readable media of claim 9, wherein: the first sensor data and the second sensor data include time series data; and the operations further comprise beginning to filter the high-frequency data at an earlier time in the time series data than a later time at which beginning to filter the low-frequency data. ### Claim 16 An autonomous vehicle comprising: a first sensor and a second sensor; one or more processors; and one or more computer-readable media storing instructions that, when executed by the one or more processors, cause the autonomous vehicle to perform operations comprising: receiving first sensor data associated with the first sensor and second sensor data associated with the second sensor; determining third sensor data by filtering high-frequency data and low-frequency data from the first sensor data and the second sensor data; determining, based at least in part on the third sensor data, a discrepancy metric associated with the first sensor; determining, based at least in part on the discrepancy metric exceeding a threshold metric, an error associated with the first sensor; and using, based at least in part on the error associated with the first sensor, the second sensor data instead of the first sensor data to control of the autonomous vehicle. ### Claim 17 (depends on 16) The autonomous vehicle of claim 16, wherein: the filtering of high-frequency data includes using a low-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor noise; the filtering of low-frequency data includes using a high-pass filter including one or more parameters configured to devalue sensor data corresponding at least in part to sensor bias; and the low-pass filter and the high-pass filter comprise one or more of: an exponential filter, a finite impulse response filter, an infinite impulse response filter, an exponential decay filter, or a median filter. ### Claim 18 (depends on 16) The autonomous vehicle of claim 16, wherein the discrepancy metric includes a first residual that is associated with the first sensor and that is higher than a second residual associated with the second sensor. ### Claim 19 (depends on 16) The autonomous vehicle of claim 16, wherein: the first sensor data and the second sensor data include time series data; and the operations further comprises beginning to filter the high-frequency data at an earlier time in the time series data than a later time at which beginning to filter the low-frequency data. ### Claim 20 (depends on 1) The method of claim 1, wherein: the first sensor and the second sensor include a first inertial measurement unit and a second IMU associated with respective positions relative to the vehicle, the method further comprises: receiving fourth sensor data associated with a third IMU; determining, by transformation to a reference point associated with the vehicle, transformed sensor data based on the first sensor data, the second sensor data, and the fourth sensor data; and the discrepancy metric is based on the transformed sensor data.