Sensor-fused flight event detection using aircraft telemetry and mobile device sensors

1 . A system for sensor-fused verification of pilot logbook entries, comprising:

a. a mobile computing device configured to be carried aboard an aircraft and comprising sensors including at least a global navigation satellite system (GNSS) receiver and an accelerometer;

b. a network interface configured to obtain aircraft telemetry for the aircraft;

c. a non-transitory memory storing airport geo-boundaries including ramp, taxiway, and runway gate polygons, event-detection thresholds, durations, tolerances, and sampling cadences; and

d. one or more processors configured to:

i. record time-stamped GNSS coordinates, ground speed, altitude, and accelerometer outputs while the aircraft traverses ground and airborne phases;

ii. automatically detect block-out, taxi, takeoff, wheels-off, wheels-on, landing rollout, and block-in by evaluating the recorded signals against the stored thresholds and gate polygons;

iii. upon detecting airborne conditions based on altitude increase and vertical-acceleration exceeding an airborne threshold for at least a first duration stored in the memory, switch sampling modalities by reducing a GNSS sampling cadence and initiating periodic retrieval of aircraft telemetry at a higher cadence, and upon detecting landing conditions based on altitude decrease and deceleration for at least a second duration stored in the memory, revert to the GNSS sampling cadence;

iv. compute event time stamps and a block time from the detected events;

v. retrieve a digital pilot logbook entry corresponding to the aircraft operation;

vi. normalize the digital pilot logbook entry by parsing time fields, time zones, and event labels into canonical fields stored in the memory;

vii. compare the computed event time stamps and durations with corresponding normalized values in the digital pilot logbook entry using the stored tolerances; and

viii. output a validation result confirming the digital pilot logbook entry when the comparison is within the stored tolerances and otherwise flagging a discrepancy together with sensor-derived evidentiary data.

2 . The system of claim 1 , wherein the block time is determined from transitions across the ramp, taxiway, and runway gate polygons stored in the memory.

3 . The system of claim 1 , wherein airborne detection comprises detecting a vertical acceleration greater than an airborne-acceleration threshold for at least the first duration in combination with ground-speed exceeding a taxi threshold, the airborne-acceleration threshold, the first duration, and the taxi threshold being stored in the memory.

4 . The system of claim 1 , wherein wheels-on detection comprises identifying a vertical-acceleration peak greater than a touchdown-impulse threshold during a detection window shorter than a third duration and followed by deceleration, and wherein the processors compute a touchdown-force metric as at least a maximum vertical acceleration within a time window centered on the detected peak, the threshold and durations being stored in the memory.

5 . The system of claim 1 , wherein the aircraft telemetry comprises Automatic Dependent Surveillance-Broadcast (ADS-B) data.

6 . The system of claim 1 , wherein the processors implement a first sampling cadence for taxi, a second sampling cadence for takeoff, approach, and landing that is greater than the first sampling cadence by at least a cadence-ratio stored in the memory, and a third sampling cadence for cruise that is less than the first sampling cadence.

7 . The system of claim 1 , wherein the processors ingest meteorological reports for departure, enroute, and destination, parse the reports to extract wind, visibility, cloud base, and temperature, classify the conditions as instrument meteorological conditions (IMC) or visual meteorological conditions (VMC) using stored rules, and determine whether instrument-flight or weather annotations in the normalized digital pilot logbook entry are consistent with the classified conditions at recorded times.

8 . The system of claim 7 , wherein the processors compute approach-stability indicators including at least a glide-path indicator derived from altitude and vertical rate over a final approach segment and a crosswind component derived from runway heading and wind, and flag an unstable approach when either indicator exceeds a corresponding threshold stored in the memory.

9 . The system of claim 1 , wherein the processors retrieve a filed flight plan and determine a route-adherence score by comparing recorded positions to the filed route within a route-proximity tolerance and a time-alignment tolerance, the tolerances being stored in the memory, and wherein the processors flag a route deviation when the score falls below a threshold stored in the memory.

10 . The system of claim 1 , wherein the processors verify pilot qualifications by accessing a pilot certification profile comprising at least aircraft type rating information and flight-condition qualification information, and flag a mismatch when the aircraft type or recorded conditions are not covered by the pilot certification profile.

11 . The system of claim 1 , wherein the processors issue a maintenance notification upon detecting a hard landing defined as the touchdown-force metric exceeding a G-force threshold stored in the memory, a bounce defined as multiple vertical-acceleration peaks within a fourth duration following wheels-on, or a landing outside a touchdown zone defined as a polygon within a runway polygon stored in the memory.

12 . The system of claim 1 , wherein the processors obtain historical aircraft telemetry by querying a repository using an aircraft identifier and a time window, compute historical wheels-off, wheels-on, and block times from the historical telemetry using the stored thresholds and durations, and reconcile the computed historical events against past digital pilot logbook entries using the stored tolerances to produce a reconciliation report.

13 . The system of claim 1 , wherein the processors render a graphical user interface on a display of the mobile computing device that presents the validation result and an interactive control to accept or dispute the result, receive a pilot attestation input via the control, and store the attestation with a timestamp and device identifier.

14 . The system of claim 1 , wherein the validation result includes a graphical user interface presenting matched flights, missed matches, and discrepancy values between reported and computed times, with selectable links to corresponding sensor traces.

15 . The system of claim 1 , wherein the processors compute altitude relative to terrain by combining GNSS altitude with terrain elevation from a terrain dataset and use the altitude relative to terrain in the wheels-off and wheels-on detections.

16 . The system of claim 1 , wherein the processors record exception states including at least one exception code selected from a set stored in the memory comprising telemetry unavailable, sensor dropout, gate-crossing ambiguity, clock skew, time-zone mismatch, and missing logbook field, and store pointers to the associated sensor traces.

17 . An onboard apparatus for sensor fused verification of pilot logbook entries, comprising:

a. a mobile computing device configured to be carried aboard an aircraft, the mobile computing device comprising a GNSS receiver, an accelerometer, a microphone interface, a camera interface, a wireless network interface configured to obtain aircraft telemetry for the aircraft, and non-volatile memory storing airport geo boundaries including ramp, taxiway, and runway gate polygons, event detection thresholds, durations, tolerances, and sampling cadences; and

b. a programmed controller executed by the mobile computing device and configured to:

i. record time-stamped GNSS coordinates, ground speed, altitude, and accelerometer outputs while the aircraft traverses the ramp, taxiway, and runway gate polygons and during airborne segments;

ii. automatically detect block-out, taxi, takeoff, wheels-off, wheels-on, landing rollout, and block-in by evaluating the recorded outputs against the stored thresholds, durations, and ramp, taxiway, and runway gate polygons;

iii. upon detecting airborne conditions based on altitude increase and vertical acceleration exceeding an airborne threshold for at least a first duration stored in the non volatile memory, switch sampling modalities by reducing a GNSS sampling cadence and initiating periodic retrieval of the aircraft telemetry at a higher cadence, and upon detecting landing conditions based on altitude decrease and deceleration for at least a second duration stored in the non volatile memory, revert to the GNSS sampling cadence;

iv. compute event time stamps and block time from detected events and compute a touchdown-force metric as a maximum vertical acceleration within a detection window about wheels-on;

v. retrieve a digital pilot logbook entry corresponding to the aircraft operation;

vi. normalize the digital pilot logbook entry by parsing time fields, time zones, and event labels into canonical fields stored in the non volatile memory;

vii. compare the computed event time stamps and durations with corresponding normalized values in the digital pilot logbook entry using the stored tolerances; and

viii. output a validation result confirming the digital pilot logbook entry when the comparison is within the stored tolerances and otherwise flagging a discrepancy together with sensor derived evidentiary data.

18 . The apparatus of claim 17 , wherein the controller, responsive to airborne detection based on altitude increase and vertical-acceleration exceeding the airborne threshold for at least a first duration in combination with ground speed exceeding a taxi threshold stored in the non volatile memory, reduces GNSS sampling cadence and initiates periodic retrieval of aircraft telemetry at a higher cadence, and responsive to landing detection based on altitude decrease and deceleration for at least a second duration and further based on identifying a vertical acceleration peak greater than a touchdown impulse threshold during a detection window shorter than a third duration stored in the non volatile memory, reverts to the GNSS sampling cadence.

19 . The apparatus of claim 17 , wherein the aircraft telemetry comprises ADS-B data.

20 . The apparatus of claim 17 , wherein the controller sets a first sampling cadence for taxi, a second sampling cadence for takeoff, approach, and landing that is greater than the first sampling cadence by at least a cadence-ratio stored in the non-volatile memory, and a third sampling cadence for cruise that is less than the first sampling cadence.

21 . The apparatus of claim 17 , wherein the controller detects a hard landing when the touchdown-force metric exceeds a G-force threshold stored in the non-volatile memory, detects a bounce when multiple vertical-acceleration peaks occur within a fourth duration following wheels-on, and determines whether wheels-on occurred outside a touchdown-zone polygon stored in the non-volatile memory.

22 . The apparatus of claim 17 , wherein the controller captures ramp imagery using the camera interface, performs optical character recognition of a tail number, and associates the recognized tail number with the computed block-in time.

23 . The apparatus of claim 17 , wherein the controller transmits a push notification to the mobile computing device upon generation of the validation result, renders a graphical user interface having an interactive control to accept or dispute the validation result, and stores any exception code with pointers to sensor traces.

24 . A non-transitory computer-readable medium storing airport geo boundaries including ramp, taxiway, and runway gate polygons, event detection thresholds, durations, tolerances, and sampling cadences, and storing instructions that, when executed by one or more processors of a system comprising a mobile computing device configured to be carried aboard an aircraft and a network interface configured to obtain aircraft telemetry for the aircraft, cause the processors to perform operations comprising:

a. recording time-stamped GNSS coordinates, ground speed, altitude, and accelerometer outputs during an aircraft operation while the aircraft traverses ground and airborne phases;

b. obtaining aircraft telemetry corresponding to the aircraft operation;

c. automatically detecting block out taxi, takeoff wheels off, wheels on, landing rollout, and block in by evaluating the recorded outputs against the stored thresholds and ramp, taxiway, and runway gate polygons;

d. upon detecting airborne conditions based on altitude increase and vertical acceleration exceeding an airborne threshold for at least a first duration stored in the non transitory computer readable medium, switching sampling modalities by reducing a GNSS sampling cadence and initiating periodic retrieval of the aircraft telemetry at a higher cadence, and upon detecting landing conditions based on altitude decrease and deceleration for at least a second duration stored in the non transitory computer readable medium, reverting to the GNSS sampling cadence;

e. computing event time stamps and a block time from the detected events;

f. retrieving a digital pilot logbook entry corresponding to the aircraft operation;

g. normalizing the digital pilot logbook entry by parsing time fields, time zones, and event labels into canonical fields stored in the non transitory computer readable medium;

h. comparing computed event time stamps and durations with normalized values in the digital pilot logbook entry using stored tolerances; and

i. outputting a validation result confirming the digital pilot logbook entry when the comparison is within the stored tolerances and otherwise flagging a discrepancy together with sensor derived evidentiary data.

25 . The non-transitory computer-readable medium of claim 24 , wherein the aircraft telemetry comprises ADS-B data.

26 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise applying relative sampling cadences including a first sampling cadence for taxi, a second sampling cadence for takeoff, approach, and landing that is greater than the first sampling cadence by at least a cadence-ratio, and a third sampling cadence for cruise that is less than the first sampling cadence.

27 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise ingesting meteorological reports, parsing the reports to extract weather parameters, classifying conditions as IMC or VMC using stored rules, and corroborating instrument-flight annotations in the digital pilot logbook entry.

28 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise retrieving a filed flight plan, applying a route-proximity tolerance and a time-alignment tolerance to derive a route-adherence score, and flagging a route deviation when the score falls below a threshold stored in memory.

29 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise detecting a hard landing when a touchdown-force metric exceeds a G-force threshold, detecting a bounce when multiple vertical-acceleration peaks occur within a fourth duration following wheels-on, detecting a landing outside a touchdown-zone polygon, and generating a maintenance notification.

30 . The non-transitory computer-readable medium of claim 24 , wherein the operations further comprise validating pilot identity using facial recognition of the pilot and a two-factor authentication challenge and recording a pilot attestation captured via an interactive control rendered on a display of the mobile computing device.