# US 12,217,330 B2 · Method and device for generating learning data for an artificial intelligence machine for aircraft landing assistance > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,217,330 B2 - Document ID: US12217330B2 - Status: Granted - Kind: B2 - Application number: 17/774,807 - Filed: 2020-11-03 - Published: 2022-12-22 - Granted: 2025-02-04 - Earliest claimed: 2019-11-07 - Assignees: Thales - Applicants: Thales - Inventors: Thierry Ganille; Guillaume Pabia; Christian Nouvel - Web page: https://patentgrep.com/patent/US12217330B2 ## CPC - G06T 11/00: Two-dimensional [2D] image generation - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/09: Supervised learning - G06T 5/50: using two or more images, e.g. averaging or subtraction - G06T 17/05: Geographic models - G06V 10/454: {Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]} - G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting - G06V 10/82: using neural networks - G06V 20/176: {Urban or other man-made structures} - G08G 5/21: located onboard the aircraft - G08G 5/54: for approach or landing - G06N 3/045: Combinations of networks - G06N 3/08: Learning methods - G06T 2207/10048: Infrared image - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93}) ## Abstract A computer-implemented method for generating labeled training data for an artificial intelligence machine is provided. The method comprises at least the steps of: defining parameters for a scenario of an aircraft approaching a runway; using the parameters of the scenario in a flight simulator to generate simulated flight data, the flight simulator being configured so as to simulate the aircraft in the approach phase and an associated autopilot; using the simulated flight data in a sensor simulator to generate simulated sensor data, the sensor simulator being configured so as to simulate a forward-facing sensor on board an aircraft and able to provide sensor data representative of information of interest of a runway, the simulated sensor data that are generated being representative of information of interest of the runway; and using the simulated flight data and the simulated sensor data to generate a ground truth, the ground truth associated with the simulated sensor data forming a pair of simulated labeled training data. ## Claims ### Claim 1 A computer-implemented method for generating labeled training data able to be used by an artificial intelligence machine implementing a deep learning image processing algorithm, the method comprising at least the steps of: defining parameters for a scenario of an aircraft approaching a runway; using the parameters of the scenario in a flight simulator to generate simulated flight data, said flight simulator being configured so as to simulate said aircraft in the approach phase and an associated autopilot; using the simulated flight data in a sensor simulator to generate simulated sensor data, said sensor simulator being configured so as to simulate a forward-facing sensor on board an aircraft and able to provide sensor data representative of information of interest of a runway, said simulated sensor data that are generated being representative of information of interest of said runway; using the simulated flight data and the simulated sensor data to generate a ground truth, said ground truth associated with the simulated sensor data forming a pair of simulated labeled training data; and wherein the step of generating the ground truth consists in: generating a first-level ground truth comprising labeling data representative of characteristics of the simulated runway, without a visibility restriction; detecting the visibility limit by identifying areas without visibility in the simulated sensor data; and correcting the first-level ground truth with the identified areas without visibility in order to generate said ground truth. ### Claim 2 (depends on 1) The method as claimed in claim 1, wherein the step of defining scenario parameters consists in defining parameters relating to at least one given runway, weather conditions and environmental conditions. ### Claim 3 (depends on 1) The method as claimed in claim 1, wherein the step of generating simulated flight data consists in simulating various approach paths of the simulated aircraft and the autopilot with the scenario parameters, and in generating simulated flight data for the various paths comprising the position (latitude, longitude, altitude) and attitude (yaw, pitch, roll) of the simulated aircraft. ### Claim 4 (depends on 1) The method as claimed in claim 1, wherein the step of generating the simulated sensor data consists in generating simulated infrared images. ### Claim 5 (depends on 1) The method as claimed in claim 1, wherein the visibility limit detection step consists in implementing a visibility limit detection algorithm on each set of simulated sensor data. ### Claim 6 (depends on 1) The method as claimed in claim 1, additionally comprising a step of generating a corrected ground truth control image by fusing each set of ground truth data with the corresponding simulated sensor data. ### Claim 7 (depends on 1) The method as claimed in claim 1, wherein the flight data and the sensor data are real data and the ground truth generation step comprises, before the first-level ground truth generation step, steps consisting in computing the positioning errors during landing in the real flight data and in taking these into consideration to generate the first-level ground truth. ### Claim 8 (depends on 1) The method as claimed in claim 1, additionally comprising a step consisting in storing the simulated labeled training data in a training database. ### Claim 9 (depends on 1) A device for generating labeled training data able to be used by an artificial intelligence machine implementing a deep learning image processing algorithm, the device comprising means for implementing the steps of the method as claimed in claim 1. ### Claim 10 (depends on 1) A computer program comprising code instructions stored on a memory of a computer causing a processor of the computer to execute the steps of the method for generating labeled training data able to be used by an artificial intelligence machine implementing a deep learning image processing algorithm, as claimed in claim 1, when said computer program is executed by the processor.