Forest inventory: Unmanned helicopters on a data collection mission

The analysis of the ecological and economic factors of a forest has a long tradition, reaching back to the early 18th century. While the economic factors have already been used to measure the wood yield for some time, there has been a global surge in the measurement of ecological factors in recent years. Particularly in the densely forested regions of Eastern Europe, a growing number of countries is beginning to take an increased interest in controlled forestry.

What, though, is actually measured within the scope of a forest inventory? The project's lead scientist, Stephan Weiss (Department of Smart System-Technologies) explains: "The diameter of a tree at breast height, the shape of the trunk and the position of the individual trees are of great interest and allow us to extrapolate the volume, quality and distribution of the wood." Currently, it is not possible to measure the shape of the tree trunk and the existence or thickness of the branches in an efficient manner, which means that these quality indicators are often lacking. According to Weiss, current practice mainly involves measuring by hand or with satellite images and airborne LIDAR data, using laser light to perform a dense scan of Earth's surface. However, these methods are inefficient and cannot provide reliable information about the quality of the wood.

"We propose the deployment of small, unmanned helicopters. By applying automatic image processing and path-planning as well as 3D reconstruction, they can autonomously fly under the tree canopy and derive the forest parameters from the 3D data," says Weiss . The research team, which consists of scientists from the AAU's Department of Smart System-Technologies, the Lakeside Labs GmbH, the Joanneum Research Forschungsgesellschaft mbH, the E.C.O. Institute for Ecology, and the environmental data company Umweltdata, is working to develop an airworthy prototype. This device is expected to navigate autonomously through managed forests, providing 3D information of sufficient density for the automatic analysis of ecological data, such as biomass indicators, tree positions, the coverage ratio of the shrub layer, the diameter at breast height, and the shape of the tree trunk.