Unmanned Aerial Vehicles (UAVs) have recently found usage in various fields, including transportation, agriculture, search and rescue, and surveillance. Despite UAVs' potential substantial economic effect, they must still be deployed in entirely autonomous commercial flights. The security of the flight is the first concern. In most cases, pilots must take control of the plane when even the most sophisticated autopilot systems choke under extreme circumstances. Methods based on artificial intelligence and adaptive controllers have shown to be effective in these unpredictable settings. However, non-determinism is another issue that they bring out. This study addresses the question of how these algorithms may be put to use. They employed an adaptive model to test the control parameter's performance before deploying it to the physical platform. The controller or optimizer might be nondeterministic or based on artificial intelligence. As a backup plan, include a method to retrieve the drone in case of failure. An aircraft model is created using a Neural Network, and the PID controller of a quad copter is optimized using a Genetic Algorithm. This integrated system transfers signals using networking.
Design of Small UAV Autopilot Systems Using Adaptive Controllers, PID Controllers, and AI
23.01.2025
763147 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
UAV longitudinal autopilot design using SLC and TECS controllers
Springer Verlag | 2024
|UAV longitudinal autopilot design using SLC and TECS controllers
Springer Verlag | 2024
|On-Line Learning Neural-Network Controllers for Autopilot Systems
Online Contents | 1995
|