🤖 AI Summary
This work addresses the cooperative search and tracking of multiple mobile ground targets in unknown, cluttered environments. Method: We propose a vision-driven set-membership estimation framework that fuses outputs from embedded vision systems—namely, object detection, semantic segmentation, and depth estimation—mounted on multiple UAVs. Leveraging pixel-level annotations, depth maps, and bounding boxes, we construct a distributed set-membership filter incorporating geometric constraints and collaborative decision-making mechanisms. Contribution/Results: To our knowledge, this is the first systematic integration of set-membership estimation into vision-based multi-UAV cooperative search. It provides guaranteed containment of both detected and potential target positions, along with conservative existence inference—without requiring prior maps or target motion models. Each UAV autonomously generates tight, robust state inclusion sets online, significantly improving search efficiency and localization reliability.
📝 Abstract
This paper addresses the problem of target search and tracking using a fleet of cooperating UAVs evolving in some unknown region of interest containing an a priori unknown number of moving ground targets. Each drone is equipped with an embedded Computer Vision System (CVS), providing an image with labeled pixels and a depth map of the observed part of its environment. Moreover, a box containing the corresponding pixels in the image frame is available when a UAV identifies a target. Hypotheses regarding information provided by the pixel classification, depth map construction, and target identification algorithms are proposed to allow its exploitation by set-membership approaches. A set-membership target location estimator is developed using the information provided by the CVS. Each UAV evaluates sets guaranteed to contain the location of the identified targets and a set possibly containing the locations of targets still to be identified. Then, each UAV uses these sets to search and track targets cooperatively.