Ultra-Wideband Localization in Forest Environments: Potential for Drone Applications

Recently, drones integrated with advanced remote sensing technologies such as GNSS, laser scanning, and cameras have been widely used in forest environments to collect forest data both manually and autonomously. However, as GNSS accuracy is usually poor under forest canopies and GNSS signals are often discontinuous and interrupted by forest canopies, high-performance SLAM algorithms based on laser scanning or camera and IMU data are usually used for navigation and localization of drones in forests. The major challenge with SLAM-based solutions is that their performance varies considerably across different forest environments. For example, laser scanning-based SLAM algorithms might fail in dense forests due to the obstruction of laser pulses by trees. As a result, SLAM technology relies heavily on the algorithm itself as well as the performance of IMU sensors used for reliable trajectory estimation.

Ultra-wideband (UWB) localization, as one of the most popular and accurate close-range positioning technologies for moving platforms, has been widely used in drone applications in indoor environments over the past decade to provide high-precision and large-scale positioning solutions. In addition to localization, UWB technology also provides communication capability between devices, making it suitable for drone swarm applications. Moreover, UWB devices are able to form a network precisely and autonomously based only on the ranging measurements between them, making UWB suitable for use in forest environments to overcome the limitations of GNSS. Our preliminary results show that decimeter-level positioning accuracy can be achieved by standalone UWB localization in a forest environment without prior knowledge of the UWB beacons or their positions. All UWB devices can be deployed randomly, and they can be rapidly and precisely discovered and included in the same local network. Therefore, UWB localization has great potential for drone applications in forests.