A feasibility study on using ViSP's 3D model-based tracker for UAV pose estimation in outdoor environments

Abstract

This paper presents a novel application of the Visual Servoing Platform's (ViSP) for small UAV pose estimation in outdoor environments. Given an initial approximation for the camera position and orientation, or camera pose, ViSP automatically establishes and continuously tracks corresponding features between an image sequence and a 3D wireframe model of the environment. As ViSP has been demonstrated to perform well in small and cluttered indoor environments, this paper explores the application of ViSP for UAV mapping of outdoor landscapes and tracking of large objects (i.e. building models). Our presented experiments demonstrate the data obtainable by the UAV, assess ViSP's data processing strategies, and evaluate the performance of the tracker.