Honors Theses
Date of Award
5-2026
Document Type
Undergraduate Thesis
Degree Name
BS
Department
Mechanical Engineering
Faculty Mentor
Carlos Montalvo, Ph.D.
Advisor(s)
Joseph Richardson, Ph.D. and Mohammadreza Kamaldar, Ph.D.
Abstract
Parafoil–payload systems present significant challenges in modeling and control due to their highly coupled, nonlinear dynamics and sensitivity to atmospheric disturbances. These systems have been thoroughly understood since the mid-1990s, but through the advent of artificial intelligence, new methods for guidance and control remain unexplored. This work implements a high-fidelity 9-DOF simulation framework which captures the multibody interactions between the parafoil, payload, and surrounding airflow, including a stochastic atmospheric model with layered winds and turbulence . Within this environment, LQR and neural network– based controllers are implemented and evaluated under identical conditions. Results show that, while classical controllers perform well in nominal environments, their effectiveness diminishes in dynamic conditions due to modeling limitations. In contrast, the neural network controller demonstrates improved adaptability and robustness, highlighting the potential of data-driven methods for autonomous parafoil guidance and precision landing applications.
Recommended Citation
Sherman, Peter, "Autonomous Guidance of 9-DOF Parafoil Systems Using LQR and NN Control" (2026). Honors Theses. 138.
https://jagworks.southalabama.edu/honors_college_theses/138
Included in
Computer-Aided Engineering and Design Commons, Navigation, Guidance, Control and Dynamics Commons, Other Aerospace Engineering Commons, Systems Engineering and Multidisciplinary Design Optimization Commons