17–23 Aug 2025
Helmholtz-Zentrum Dresden-Rossendorf (HZDR)
Europe/Berlin timezone

Computational Modeling of Aircraft Flight and Machine Learning-Based Control

Not scheduled
20m
Room 102 (Building 270, HZDR)

Room 102

Building 270, HZDR

Description

In this presentation, I will talk about general computational flight mechanics, general methods of simulating flight using different tools like Python3 (simulations and mathematical modelling), OpenVSP (CFD), and Python libraries like SciPy and NumPy. I will explain how these tools can be used to model aircraft and simulate flight, using the most basic numerical methods. Furthermore, I will cover how the data generated from these simulations can be utilized to train machine learning models for autonomous flight control, using frameworks like PyTorch or TensorFlow. I will also talk about my goals and plans for this research, including future steps like simulating real 3D flight and adding image processing. I plan to apply these methods to real-life situations, especially for controlling unmanned aerial vehicles (UAVs). This work demonstrates the potential of combining physics-based modeling with machine learning to advance autonomous flight.

Primary author

Luka Guraspashvili (Georgian Technical University(GTU))

Presentation materials