Autonomous navigation of Unmanned Aerial Vehicles (UAVs) for border patrol: A stochastic framework

Abstract

This study focuses on the use of Unmanned Aerial Vehicles (UAVs) in internal security operations, particularly border patrolling. The primary objective is to develop a stochastic navigation strategy for UAVs that maximises the mission success rate in the face of uncertainty in target movement. Traditional deterministic UAV navigation strategies are susceptible to exploitation by intelligent adversaries who can learn and predict the UAVs' routes. To address this vulnerability, we introduce a probabilistic approach to UAV navigation, where the UAVs' movements are governed by stochastic processes, making their paths unpredictable and thus more resilient to counter-strategies. UAV and target movements are modelled through simulation, and a fully autonomous UAV algorithm is developed by testing several derivative-free optimisation methods: Simulated Annealing algorithm as a heuristic approach, a simplex-based Stochastic Nelder-Mead method, and Radial Basis Function (RBF) as an interpolation-based Response Surface Methodology. These algorithms enable the UAV to autonomously search, pursue, and defend the area against targets and are tested through simulation in border violation scenarios. Experiments with the algorithms demonstrate that RBF outperforms the others in terms of objective function value and execution time. By employing RBF, the movement strategy of the UAV is optimised, striking a balance between exploration and exploitation of the search space. The results showcase the effectiveness of the proposed stochastic navigation approach in scenarios where limited information about the target's movement is available. The mathematical tools and frameworks developed in this study are applicable to various real-life contexts, including disaster management, search and rescue operations, and defence and security missions. This research makes a significant contribution to the field of surveillance technologies by presenting a sophisticated approach to enhance UAV navigation in complex and uncertain environments. By introducing stochastic elements into the UAVs' movement patterns, this study addresses a critical vulnerability in traditional deterministic navigation strategies, ultimately improving mission success rates, reducing mission completion time, and enhancing the overall robustness and adaptability of UAV-based security systems.

More information

Identifiers

ORCID for Joerg Fliege: orcid.org/0000-0002-4459-5419

ORCID for Toni Martinez Sykora: orcid.org/0000-0002-2435-3113

Catalogue record

Export record