Interaction control with an aerial robot
Loading...
Date
2025-06-16
Authors
Τσακίρη, Δήμητρα
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis presents a comprehensive approach to the control, simulation, and visualization of aquadcopter equipped with advanced mechanisms for position and force regulation, with a specificemphasis on dynamic interaction with a deformable surface. The primary objective of this studyis to develop an efficient and adaptive control framework that enables the quadcopter to follow apredefined trajectory while exerting a controlled normal force on the surface via an attached probe.To achieve this, a dynamic model of the aerial vehicle and the surface is constructed to accuratelycapture the behaviors and coupling effects that emerge during contact. The simulation involves theimplementation and testing of three different control strategies. First, we present the developmentof a Proportional-Integral-Derivative (PID) controller, second, of a neuro-adaptive controller, andthird, of a more advanced final controller that combines the two previous strategies and consistsof two stages, one for position and one for force control. The controllers enable the drone tonavigate from an initial position to a desired position, while maintaining stability and accuratecontrol over its trajectory and altitude. To enhance robustness, a Kalman filter is employed to fusenoisy sensor measurements and provide real-time state estimates, improving positioning precision.These estimates are used as feedback to the controller to generate control inputs for the drone,ensuring smoother and more stable flight. The program also simulates surface deformations inresponse to the drone’s interactions, providing a realistic portrayal of physical effects such as orcedistribution and height adjustments. The surface adapts in real time to reflect forces applied bythe drone, visualized in the animation. The inclusion of drone-surface interaction is a significantenhancement, as it enables the simulation of scenarios relevant to various real-life applications.By examining deformation and force response, this study aims to support the development ofdrones capable of performing precise tasks such as sampling, or inspecting surfaces in hazardous or inaccessible environments. In conclusion, this thesis paves the way for further optimization and research in UAV applications. The integration of intelligent UAVs with advanced sensing systems has the potential to significantly enhance both civilian and military operations by improving target acquisition and situational awareness tasks.
Description
Keywords
Control interaction, Robots, Quadcopter