BiodivKI
SmartPatrol
More efficient against poaching through AI: dynamic real-time optimization of protected area patrols
Poaching is an acute threat to the biodiversity of terrestrial wildlife and the ecosystems that depend on them. An important intervention against this is ranger patrols to remove wire snares. Due to logistical difficulties and high costs, new strategies are needed to plan more efficient patrols.
The project aims to use AI to enable near-real-time analysis of biodiversity and patrol data and the optimization of patrols based on this. To this end, an AI model is to be developed that automatically identifies species on camera trap photos and integrates them into an automated workflow that analyzes the data. In a second workflow, patrol data is cleaned up using AI. The result of both workflows creates the basis for AI-based formal optimization of future patrols to protect biodiversity. In cooperation with local conservation organizations, the solution is being tested in Vietnam, where many endemic species are found and where there is a high risk of poaching.
In the first phase, the project pursues the in-depth development of a concept and the preparation of an application for a second phase, which focuses on the implementation of the aforementioned objectives. In addition, initial scientific work will be carried out in the project to allow an effective and rapid entry into a possible second phase.