arXiv:2512.09898v1 Announce Type: cross 
Abstract: The integration of Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) is increasingly central to the development of intelligent autonomous systems for applications such as search and rescue, environmental monitoring, and logistics. However, precise coordination between these platforms in real-time scenarios presents major challenges, particularly when external localization infrastructure such as GPS or GNSS is unavailable or degraded [1]. This paper proposes a vision-based, data-driven framework for real-time UAV-UGV integration, with a focus on robust UGV detection and heading angle prediction for navigation and coordination. The system employs a fine-tuned YOLOv5 model to detect UGVs and extract bounding box features, which are then used by a lightweight artificial neural network (ANN) to estimate the UAV's required heading angle. A VICON motion capture system was used to generate ground-truth data during training, resulting in a dataset of over 13,000 annotated images collected in a controlled lab environment. The trained ANN achieves a mean absolute error of 0.1506{\deg} and a root mean squared error of 0.1957{\deg}, offering accurate heading angle predictions using only monocular camera inputs. Experimental evaluations achieve 95% accuracy in UGV detection. This work contributes a vision-based, infrastructure- independent solution that demonstrates strong potential for deployment in GPS/GNSS-denied environments, supporting reliable multi-agent coordination under realistic dynamic conditions. A demonstration video showcasing the system's real-time performance, including UGV detection, heading angle prediction, and UAV alignment under dynamic conditions, is available at: https://github.com/Kooroshraf/UAV-UGV-Integration

تم اقتراح إطار عمل جديد لدمج الطائرات بدون طيار (UAV) والمركبات الأرضية غير المأهولة (UGV) يركز على التنقل والتنسيق في الوقت الحقيقي من خلال الكشف القوي عن UGV وتوقع زاوية الاتجاه باستخدام نموذج YOLOv5 المعدل وشبكة عصبية صناعية خفيفة. يعالج هذا التطور تحديات كبيرة في السيناريوهات التي يكون فيها نظام تحديد المواقع العالمي (GPS) أو نظام تحديد المواقع العالمي عبر الأقمار الصناعية (GNSS) غير متاح.

Se ha propuesto un nuevo marco para la integración de vehículos aéreos no tripulados (UAV) y vehículos terrestres no tripulados (UGV), centrado en la navegación y coordinación en tiempo real mediante la detección robusta de UGV y la predicción del ángulo de dirección utilizando un modelo YOLOv5 ajustado y una red neuronal artificial ligera. Este desarrollo aborda desafíos significativos en escenarios donde el GPS o el GNSS no están disponibles.

Un nouveau cadre pour l'intégration des véhicules aériens sans pilote (UAV) et des véhicules terrestres sans pilote (UGV) a été proposé, axé sur la navigation et la coordination en temps réel grâce à une détection robuste des UGV et à la prédiction de l'angle de cap à l'aide d'un modèle YOLOv5 affiné et d'un réseau de neurones artificiels léger. Ce développement répond à des défis importants dans des scénarios où le GPS ou le GNSS est indisponible.

A new framework for integrating Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) has been proposed, focusing on real-time navigation and coordination through robust UGV detection and heading angle prediction using a fine-tuned YOLOv5 model and a lightweight artificial neural network. This development addresses significant challenges in scenarios where GPS or GNSS is unavailable.

Visual Heading Prediction for Autonomous Aerial Vehicles

Was this article worth reading? Share it

LucidQuery AI

Vermeer

Attentive AI

Postugc

Dyad

FETCH HIVE

Ready to build your own newsroom?