Liujun Li

This work presents an AI-driven modularized low-cost spot spraying system that uses threshold-based decision-making to dynamically adjust herbicide application in real-time. The system integrates OAK-D for weed detection, Jetson Orin Nano for plant classification, and Arduino Mega for spray control. Simulations demonstrate that this approach significantly reduces herbicide consumption, improves spray precision, and effectively adapts to varying environmental conditions, making it a scalable solution for precision agriculture.

UAVs face some critical challenges in pervasive applications, such as navigation in GPS-denying areas, obstacle avoidance, and tradeoff between payload and flight time. The INSPIRE UTC at Missouri S&T is developing a clamping UAV, a hybrid flying and traversing robot, to make it possible that the UAV flies under a bridge girder, grabs the girder, and moves along the girder for inspection. This heavy UAV is difficult to control underneath the deck when equipped with multiple cameras. In this study, CFD modeling and simulation are conducted to understand the flight behavior and ceiling effect of the UAV under the bridge deck. First, 3D scanning and reverse engineering are used to build CFD models. Second, a series of single propeller tests are conducted to validate the process of CFD modeling and the simulation results. Third, based on the workflow and technique validated through single propeller CFD analysis, the CFD model of an entire UAV is established and analyzed to predict the behavior of the UAV and understand its interaction with upper boundaries as it approaches the bridge ceiling vertically. The CFD simulation results show that the ceiling effect of the designed clamping UAV is insignificant when the UAV approaches the standard bridge deck with enough depth. These results provide a technical reference for the design and control of the clamping UAV for bridge inspection.