Unmanned aerial vehicle (UAV) swarms are essential for mission-critical aerial Internet of Things (IoT) applications. However, reliable multi-hop routing in these swarms is severely challenged by highly dynamic topologies and partial, asynchronous observations. Furthermore, conventional training data poorly represent rare but decisive disruptions, such as terrain occlusion, no-fly-zone detours, and congestion bursts. To address these issues, this paper proposes a deployment-aligned robust learning framework organized into three functional layers. At the policy layer, we propose the Multi-Agent Opportunistic (MAO) routing method, learned via a Belief Graph-based MAPPO (BG-MAPPO) algorithm. To overcome partial and stale observations, BG-MAPPO constructs a local belief graph encoding the dynamic neighborhood state, which is then used to jointly determine the optimal forwarder-set size and a diversity-aware routing distribution. At the training layer, the policy is optimized inside a terrain-aware, physics-calibrated Digital Twin Network (DTN). To prevent overfitting to limited training distributions, a diffusion-based generator (FD3M) enriches the training data with trajectory-and-flow samples covering critical corner cases. At the validation layer, large-scale simulations show that MAO reduces end-to-end delay by 27–34% compared to representative baselines. The policy also incurs modest operational overhead, maintaining millisecond-level inference and bounded signaling. Finally, hardware-in-the-loop (HIL) experiments confirm that the DTN accurately predicts physical execution, with relative gaps of only 4% in packet delivery ratio and 7% in delay, effectively bridging the sim-to-real gap for robust UAV-swarm networking.

Path planning is a fundamental application of an unmanned aerial vehicle (UAV) swarm. Performing such a task in a complex mountain environment with a wind field would encounter challenges such as the sim-to-real (simulation-to-reality) gap. In this paper, we present a digital twin (a virtual replica of the physical system)-assisted path-planning framework for a UAV swarm with two phases: global path planning and real-time trajectory planning. For global path planning, we develop an optimization model that combines the constraints of a single UAV and the swarming rules, while also accounting for wind effects. To solve the optimization model, we improve the polar lights optimization (PLO) algorithm via multiple strategies (named PLOM), enhancing the initialization, exploitation, exploration, and equilibration processes. The major improvement strategies consist of opposition-based learning with refraction and elite, the logarithmic spiral motion, the Cauchy-Gaussian operator, and sine cosine perturbation. We construct a realistic geographical simulation environment based on a digital elevation model (DEM) and dominant wind, and design controlled experiments under different conditions of UAVs, threats, waypoints, and wind speeds. The simulation results demonstrate that the PLOM algorithm always achieves the best solution in swarm path planning scenarios with different complexities. Meanwhile, the PLOM algorithm has the greatest robustness with nearly the shortest runtime.

With the rapid development of low-altitude economies, unmanned aerial vehicle (UAV) swarm has attracted growing interest for cooperative target search. However, most existing studies focus on static targets and assume UAVs operate at a single horizontal altitude, limiting their practical applicability. This paper proposes a novel multi-UAV cooperative search framework for moving targets based on multi-agent deep reinforcement learning (MADRL). By coordinating UAVs across high, medium, and low-altitude layers, the system achieves improved search efficiency through altitude-adaptive operations. We further introduce a revisit-time compensation mechanism to enhance detection performance for moving targets in a multi-layer UAV swarm. To address the challenges of slow convergence and sparse feedback in MADRL, we propose hybrid-reward-based value decomposition networks (HRVDN) algorithm that integrates dense local rewards with sparse global rewards, accelerating learning while encouraging agents to collect high-value information. Simulation results demonstrate that the proposed approach outperforms existing methods in terms of target search rate and area coverage.

Digital twin network (DTN) technology provides significant support for intelligent applications of unmanned aerial vehicle (UAV) swarms. However, related research focuses on DTN applications and lacks attention to the construction and maintenance of high-fidelity digital twin (DT) models. In this paper, we first develop a DT simulation platform for UAV swarms. Then, a dynamic data-driven DT model calibration scheme is proposed on the example of the most fundamental flight state of a UAV. The scheme utilizes parameter identification to estimate offline the key parameters of the measurement model and system deviations. Furthermore, the flight state is corrected online utilizing data assimilation actuated on the actual and virtual data. Simulation experiments on the DT simulation platform are conducted, and the effects of data sampling rate on calibration accuracy and computational load are analyzed. The results demonstrate that parameter identification and data assimilation significantly improve the fidelity of the DT model in terms of the optimal sub-pattern assignment (OSPA) metric to different degrees.

As an advanced framework, the digital twin network (DTN) provides effective management and decision support for autonomous aerial vehicle (AAV) swarms and has become a recent research hotspot. The effectiveness of many DTN applications relies on the assumption that high-fidelity digital twin (DT) models exist and are readily available. However, constructing such high-fidelity DT models of AAV swarms is a challenging task, especially in complex and dynamic environments. Despite its importance, there is a notable lack of research focused on the construction of high-fidelity DT models specifically for AAV swarms. This study proposes a dynamic data-driven approach for constructing and calibrating DT models of AAV swarms to achieve long-term consistency with real-world AAV behaviors. The method leverages parameter identification to estimate key parameters of DT models and data assimilation to refine and calibrate the model. It can provide high-fidelity DT AAV models for artificial intelligence model training and facilitate AAV swarm DTN from concept to real application. Additionally, this article developed a DT simulation platform for AAV swarms, validating the proposed method through software-in-the-loop simulations and physical testing. Results indicate that the optimal subpattern assignment metric decreases by an average of 79.2% after calibration, significantly improving the DT model’s fidelity.

Digital Twin (DT) technology represents a cutting-edge methodology that digitally maps physical entities with high fidelity, leading to the Digital Twin Network (DTN) through its integration with network technologies. DTN establishes bidirectional communication between virtual and physical spaces, enabling real-time monitoring, dynamic optimization, and precise control of physical networks. This addresses challenges posed by network expansion and service diversification, revolutionizing the management and optimization of complex network systems. Despite its potential, DTN implementation remains challenging, with research still nascent and lacking detailed guidelines. This paper aims to bridge this gap by presenting a comprehensive survey of the reference architecture for real-world DTN implementation and its key enabling technologies. It begins by defining the conceptual foundation of DTN and reviewing related architectural studies. This is followed by the proposal of a universal and scalable modular DTN architecture, encompassing the physical layer, data layer, DT model layer, and service layer. We then explore the critical enabling technologies required for implementing this architecture and analyze applications enhanced by DTN. Notably, We propose a five-level digital twin model evolution taxonomy framework that systematically reveals the evolution path from basic mapping to ultra-high-fidelity autonomous inference. This framework provides a structured evaluation benchmark for optimizing and advancing digital twin models. Finally, we discuss the primary open issues in DTN, offering theoretical and practical guidance for future research in this field.

Cooperative target search (CTS) technology is highly desirable in various multi-autonomous aerial vehicle (AAV) applications. However, searching for unknown targets in a dynamic threatening environment is a challenging problem, especially for AAVs with limited sensing range and communication capabilities. Besides, traditional searching methods lack scalability and efficient collaboration among the AAV swarm in dynamic environments. In this work, a digital twin (DT)-enabled distributed CTS approach was presented for AAV swarms and achieving sim-to-real transfer. Specifically, a new scalable multi-agent reinforcement learning (MARL) based algorithm called SAMARL is adopted to improve effectiveness and adaptability, combining a multi-head attention mechanism. In SAMARL, a scalable observation space with graph representation and an environmental cognition map is designed to thoroughly consider the target search rate, area coverage, and safety assurance. Then, a DT-driven training framework is proposed to facilitate the continuous evolution of MARL models and address the tradeoff between training speed and environment fidelity. Furthermore, we innovatively develop a distributed AAV swarm digital twin cooperative target search validation system, including real flight control, communication simulation tools, and a 3D physics engine. Extensive simulations validate its superiority compared to state-of-the-art strategies. More importantly, we also conduct real-world flight experiments on different scale mission areas and AAV swarms, further demonstrating the generalization and scalability of trained models.

Over the past decades, unmanned aerial vehicles (UAVs) have been widely used in both military and civilian fields. In these applications, flocking motion is a fundamental but crucial operation of multi-UAV systems. Traditional flocking motion methods usually designed for a specific environment. However, the real environment is mostly unknown and stochastic, which greatly reduces the practicality of these methods. In this article, deep reinforcement learning (DRL) is used to realize the flocking motion of multi-UAV systems. Considering that the sim-to-real problem restricts the application of DRL to the flocking motion scenario, a digital twin (DT)-enabled DRL training framework is proposed to solve this problem. The DRL model can learn from DT and be quickly deployed on the real-world UAV with the help of DT. Under this training framework, this article proposes an actor–critic DRL algorithm, named behavior-coupling deep deterministic policy gradient (BCDDPG), for the flocking motion problem, which is inspired by the flocking behavior of animals. Extensive simulations are conducted to evaluate the performance of BCDDPG. Simulation results show that BCDDPG achieves a higher average reward and performs better in terms of arrival rate and collision rate compared with the existing methods.

Directional antennas have numerous advantages over omnidirectional antennas in CSMA/CA ad hoc networks. However, estimating the available bandwidth of a flow in the medium access process in such a network is very challenging. In this paper, we present a passive available bandwidth estimation algorithm for directional ad hoc networks, termed PABE-D. The estimation process consists of two phases, i.e., the preliminary and refined estimation phases. In the preliminary phase, the available transmission/reception duration in each beam of the node and the available bandwidth of the directional link are obtained by analyzing the available duration of both the sender and receiver. In the refined phase, we analyze the effect of the directional hidden terminal and deafness problems to further improve the estimate accuracy. The performance of our proposed PABE-D algorithm is evaluated in two typical topologies, i.e., the parallel and random grid topologies. The simulation results demonstrate that our algorithm can effectively estimate the available bandwidth in directional ad hoc networks.