WildGraph: Realistic Graph-based Trajectory Generation for Wildlife
Research Poster Engineering 2025 Graduate ExhibitionPresentation by Ali Al Lawati
Exhibition Number 21
Abstract
Trajectory generation is an important task in movement studies; it circumvents the privacy, ethical, and technical challenges of collecting real trajectories from the target population. In particular, real trajectories in the wildlife domain are scarce as a result of ethical and environmental constraints of the collection process. In this paper, we consider the problem of generating long-horizon trajectories, akin to wildlife migration, based on a small set of real samples. We propose a hierarchical approach to learn the global movement characteristics of the real dataset and recursively refine localized regions. Our solution, WildGraph, discretizes the geographic path into a prototype network of H3 (this https URL) regions and leverages a recurrent variational auto-encoder to probabilistically generate paths over the regions, based on occupancy. WildGraph successfully generates realistic months-long trajectories using a sample size as small as 60. Experiments performed on two wildlife migration datasets demonstrate that our proposed method improves the generalization of the generated trajectories in comparison to existing work while achieving superior or comparable performance in several benchmark metrics.
Importance
Wildlife trajectories are used for various conservation tasks, including anti-poaching. However, collecting real trajectory data is challenging and presents risks to both wildlife and researchers. In this work, we propose a method that learns the characteristics of a limited set of real trajectories and generates synthetic ones that closely resemble them. Our approach facilitates the training of advanced learning models, enhances simulation tasks, and supports applications that require larger datasets for improved performance.
DEI Statement
Our work contributes to wildlife conservation by enabling scientists to model animal movement using a limited set of real-world samples. Our approach promotes wildlife conservation by helps identify critical patterns, such as poaching hotspots. This work supports Diversity, Equity, and Inclusion (DEI) by empowering economically disadvantaged communities with accessible tools to promote wildlife protection and mitigate threats to at-risk species.