Bionote
Jacob Swindell is a first-year PhD student at the University of Lincoln. His current research focuses on efficient, informed, adaptive path planning for UAVs and is funded as part of the AgriFoRwArdS CDT. Jacob hopes to improve the speed and quality of UAV aerial surveys, allowing farmers to receive more information when making vital management decisions. Jacob’s broad research interests include simulation design, probabilistic machine learning and UAV path planning algorithms. Jacob has a BSc in Computer Science and an MSc in Robotics from the University of Lincoln.
Presentation Abstract
Accurate agricultural weed mapping using UAVs is crucial for precision farming applications. Traditional methods rely on orthomosaic stitching from rigid flight paths, which is computationally intensive and time-consuming. Gaussian Process (GP)-based mapping offers continuous modelling of the underlying variable (i.e. weed distribution) but requires discretisation for practical tasks like path planning or visualisation. Current implementations often default to quadtrees or gridmaps without systematically evaluating alternatives. This study compares five discretisation methods: quadtrees, wedgelets, top-down binary space partition (BSP) trees using least square error (LSE), bottom-up BSP trees using graph merging, and variable-resolution hexagonal grids. Evaluations on real-world weed distributions measure visual similarity, mean squared error (MSE), and computational efficiency. Results show quadtrees perform best overall, but alternatives excel in specific scenarios: hexagons or BSP LSE suit fields with large, dominant weed patches, while quadtrees are optimal for dispersed small-scale distributions. These findings highlight the need to tailor discretisation approaches to weed distribution patterns (patch size, density, coverage) rather than relying on default methods. By choosing representations based on the underlying distribution, we can improve mapping accuracy and efficiency for precision agriculture applications.
Co-Researcher (Supervisor): Riccardo Polvara