Author

Date of Award

Fall 2025

Document Type

Open Access Thesis

Department

Mechanical Engineering

First Advisor

Junsoo Lee

Abstract

Unmanned aerial vehicles (UAVs) are increasingly used in precision agriculture, where extended autonomous operation is required for monitoring, intervention, and field management. However, achieving long-term autonomy remains challenging due to battery constraints, environmental uncertainty, and the need to balance exploration with event-driven tasks. To address these challenges, a multi-layer decision-making framework inspired by Dual Process Theory (DPT) is developed. The framework combines reactive return-tobase strategies, exploratory navigation, and directional bias from prior missions, with a conflict-monitoring mechanism that adapts system behavior based on real-time conditions. The approach is implemented in a simulated agricultural grid environment, demonstrating improved adaptability and coverage under stochastic event occurrence and energy limitations. Building upon this foundation, a learning-based extension is proposed to enable UAVs to refine decision strategies over repeated missions, advancing toward scalable and robust autonomy for agricultural applications.

Rights

© 2025, Shruti Jadhav

Share

COinS