A Heterogeneous UAV UGV Methodology for Aerial Tracking of Mini Autonomous Cars

Santiago Fernandez, Angel da Silva, Hiago Sodre, Pablo Moraes, Bruna Gutterres, Ricardo B. Grando · N/A · 2026

Summary

Aerial tracking of autonomous ground vehicles can provide useful situational awareness in environments where fixed external perception infrastructure is unavailable.

Abstract Summary

Aerial tracking of autonomous ground vehicles can provide useful situational awareness in environments where fixed external perception infrastructure is unavailable. However, visual-only tracking from an unmanned aerial vehicle (UAV) is limited by camera field of view, viewpoint changes, occlusions, and temporary loss of target visibility. This paper presents a heterogeneous UAV--UGV methodology for aerial tracking of small-scale autonomous ground vehicles in simulation. The system is implemented in ROS~2 and Gazebo Sim using a Tello-like UAV equipped with a front-facing RGB camera and multiple LIMO-like unmanned ground vehicles (UGVs). Three approaches are evaluated: a training-free color-marker baseline, a YOLO-based visual baseline, and a proposed UAV--UGV fusion method. The proposed method uses the same visual detector as the YOLO baseline, but combines visual detections with UGV state information to maintain the target estimate during visual gaps. The methods are evaluated across four scenarios: single-car tracking, crossing trajectories, occlusion, and UAV-assisted awareness. Results over 600 simulation runs show that the proposed fusion method achieves nearly continuous track coverage across all scenarios while reducing localization error relative to the YOLO baseline. Compared with the visual-only YOLO approach, the fusion method reduces RMSE by 37.3\%, 64.2\%, 62.6\%, and 61.8\% in the four evaluated scenarios, respectively. The method also enables conflict-awareness lead times of 10.20~s in the crossing scenario and 9.07~s in the UAV-assisted awareness scenario. These results indicate that heterogeneous UAV--UGV fusion can improve aerial tracking continuity and robustness when visual observations are intermittent.

Key Points

Aerial tracking of autonomous ground vehicles can provide useful situational awareness in environments where fixed external perception infrastructure is unavailable.
However, visual-only tracking from an unmanned aerial vehicle (UAV) is limited by camera field of view, viewpoint changes, occlusions, and temporary loss of target visibility.
This paper presents a heterogeneous UAV--UGV methodology for aerial tracking of small-scale autonomous ground vehicles in simulation.
The system is implemented in ROS~2 and Gazebo Sim using a Tello-like UAV equipped with a front-facing RGB camera and multiple LIMO-like unmanned ground vehicles (UGVs).
Three approaches are evaluated: a training-free color-marker baseline, a YOLO-based visual baseline, and a proposed UAV--UGV fusion method.

Methodology

The proposed methodology (Figure 1 ) is implemented in ROS 2 Jazzy and Gazebo Sim. The simulated heterogeneous team consists of a mini UAV and one or more UGVs. The UAV is equipped with a front-facing RGB camera and observes the UGVs from an elevated viewpoint. The UGVs follow predefined trajectories according to the selected scenario, while the UAV provides aerial visual information.

The simulation considers a Tello-like UAV and an Agilex-like UGV.

Methodology — A Heterogeneous UAV UGV Methodology for Aerial Tracking of Mini Autonomous Cars — Fig. 1 : Overview of the proposed heterogeneous UAV–UGV aerial tracking methodology. The UAV provides front RGB images for visual detection, while UGV state information is used by the proposed fusion method to maintain target estimates during visual gaps. The same simulation environment is used to evaluate the color baseline, the YOLO baseline, and the proposed UAV–UGV fusion approach. Code availability: https://github.com/ricardoGrando/limo_tello_jazzy_spawn . Video: https://www.youtube.com/watch?v=f7xAwTxisKk

Results

All three methods completed the full evaluation matrix. Each method produced 200 valid runs, with 50 runs per scenario. The YOLO baseline and the proposed fusion method loaded the trained YOLO model correctly in all runs, with no fallback to the color proxy detector. No false positives or ID switches were observed. Table II summarizes the evaluation validity. Table III presents the full quantitative comparison.

The color baseline provides a simple visual reference, but its localization error is high.

Results — A Heterogeneous UAV UGV Methodology for Aerial Tracking of Mini Autonomous Cars — (a) S1: single-car tracking.

Sources and Demonstrations

Figures are reproduced from Fernandez et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.