Ultra-Wideband-Based Method for Measuring Tree Positions With Decimeter-Level Accuracy Under a Forest Canopy

This study presents a low-cost method for measuring tree positions under a forest canopy with decimeter-level accuracy using standalone ultra-wideband data. The proposed method consists of six beacons and one tag, which enable the creation of a sensor network and the localization of trees over an area of approximately 6500 m<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>. Two efficient signal transceiver mechanisms were developed for raw data collection, followed by an optimized method for setting up the beacon network. Each tree was measured twice at two distinct points around the stem. The tag's trajectory was estimated using an extended Kalman filter with non-line-of-sight processing, and the optimal position of each tree was then extracted from this trajectory. To evaluate the accuracy of the developed method, field tests were conducted in a boreal forest zone in Evo, Finland. Experimental results show that the developed method measured tree positions with a root-mean-square-error of 0.13 m compared to the total station-based reference and 0.14 m compared to the high-density airborne laser scanning-based reference measurements. Therefore, the developed method appears to be suitable for forest surveying tasks that require decimeter-level positional accuracy under a forest canopy.