Multi UAVs Autonomous Missions. Navigation, Execution Control, and Exploration

Arias Pérez, Pedro · 2025 · Crossref

DOI: 10.20868/upm.thesis.90045

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This doctoral thesis addresses the challenge of achieving full autonomy in multi-UAV systems, which currently rely heavily on teleoperation or semi-autonomous modes due to difficulties in navigating complex environments, adapting missions in real-time, and coordinating multiple agents. The research is motivated by the need to reduce human intervention in applications such as industrial infrastructure inspection and large-scale exploration. The work is guided by three specific research questions: how to enable reliable autonomous navigation in complex settings, how to execute flexible high-level missions, and how to explore unstructured environments using minimal sensing capabilities. To address these challenges, the author developed a modular, reusable software framework integrated into Aerostack2, an open-source aerial robotics architecture. The methodology involved designing three core components: robust navigation modules for complex environments, a dynamic task execution system for distributed multi-UAV operations, and a novel exploration strategy for nano-drones equipped with only minimal onboard sensors. The system was validated through extensive simulated experiments and real-world field tests. These included autonomous inspections of wind turbines and photovoltaic plants, as well as cooperative exploration missions using nano-drones in unknown, unstructured environments. The results demonstrate the framework’s scalability, robustness, and effectiveness across diverse scenarios. In industrial inspections, the system successfully executed complex frontal and back-oblique inspection missions for wind turbines and performed both area coverage and on-demand panel inspections for solar parks using multi-UAV swarms. For exploration, the minimal-sensing strategy enabled nano-drones to effectively map unstructured environments. Experimental data from simulations and real-world flights confirmed that the system maintains performance as the number of UAVs increases, validating the efficiency of the proposed frontier allocation heuristics and collision avoidance mechanisms. The significance of this work lies in its contribution to the realization of fully autonomous UAV systems. By extending Aerostack2 with advanced navigation, mission planning, and lightweight exploration capabilities, the thesis provides a solid foundation for future advancements in multi-agent coordination and real-world UAV deployment. The open-source nature of the framework facilitates further research and technology transfer, addressing the fragmentation in current aerial robotics software and enabling developers to focus on innovation rather than reinventing core functionalities. This approach enhances the interoperability and reliability of UAVs in complex, dynamic environments.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• driverless ads