US 12,136,277 B2
Collection, processing, and output of flight information method, system, and apparatus
Mahdi Al-Husseini, Waialua, HI (US); Joshua Barnett, Stanford, CA (US); Anthony Chen, Stanford, CA (US); and Joseph Divyan Thomas, West Lafayette, IN (US)
Filed by Aura Training Systems, Inc., Waialua, HI (US)
Filed on Dec. 7, 2021, as Appl. No. 17/544,787.
Claims priority of provisional application 63/122,923, filed on Dec. 8, 2020.
Prior Publication US 2022/0180647 A1, Jun. 9, 2022
Int. Cl. G06V 20/59 (2022.01); G06V 10/75 (2022.01); G06V 10/82 (2022.01); G06V 20/64 (2022.01)
CPC G06V 20/59 (2022.01) [G06V 10/751 (2022.01); G06V 10/82 (2022.01); G06V 20/64 (2022.01)] 19 Claims
OG exemplary drawing
 
1. An apparatus to obtain flight data for pilot flight maneuver training through visual analysis of a flight instrument comprising:
a computer-vision-implementing camera with sightlines to a part of an instrument panel that includes the flight instrument in the aircraft cockpit and having an onboard real time flight information machine learning chipset, a computer processor a memory, and a runtime data capture module in the memory,
wherein to obtain the flight data for the pilot flight maneuver training through visual analysis of the flight instrument, the computer-vision-implementing camera processor is to execute the runtime data capture module to obtain an image of the flight instrument, to process the image with an object detection neural network to identify the flight instrument in the image and perform a perspective transformation of the image using one or more fiducial markers in the image, to determine the flight data from the flight instrument in the image and to thereby obtain the flight data through real time device-implemented visual analysis of the flight instrument without an electrical signal from the flight instrument and without an electrical signal from any flight data recorder,
wherein the flight data are obtained from real flight environment, the flight data are used for training the object detection neural network comprising time-series data of real flight environment comprising real flight instrument, the flight data comprises of standard flight maneuver data, obtained from real flight environment, to be compared with the flight data for the pilot flight maneuver training.