Dynamic decision-making of UAV swarm based on constrained multi-objective optimization under incomplete interference information

UAV swarms have shown immense potential for applications ranging from disaster response to military reconnaissance, but ensuring reliable communication in contested environments has remained a critical challenge. These systems depend heavily on open wireless channels, making them vulnerable to external interference and malicious eavesdropping. Traditional resource allocation methods rely on perfect channel state information and complete interference data—assumptions that rarely hold in real-world scenarios where spectrum conditions shift rapidly and environmental information remains fragmentary.

Professor Lin's team at Harbin Engineering University tackled this challenge by developing an intelligent decision-making framework that fundamentally changes how UAV swarms manage spectrum resources under uncertainty. Published in the Chinese Journal of Aeronautics (https://doi.org/10.1016/j.cja.2025.103846), this innovative approach integrates fuzzy logic, dynamic multi-objective optimization, and transfer learning to enable autonomous spectrum allocation even when information is incomplete and environments are constantly evolving.

The framework addresses the inherent uncertainty of external interference through a sophisticated fuzzy logic system. Rather than requiring precise interference measurements, the system dynamically models interference impacts using fuzzy set theory, defining interference intensity, range, and spatial distance through inference rules and membership functions. This transforms vague environmental perception into actionable spectrum constraints that adapt in real-time during iterative assessment—effectively converting the unquantifiable into operational decisions.

At the heart of the system lies a dynamic constrained multi-objective optimization model that balances competing priorities simultaneously. The framework minimizes self-interference within the UAV cluster while suppressing potential eavesdropper throughput, governed by constraints including spectrum utilization rates, frequency conflict limits, and minimum communication requirements. This dual-objective approach identifies optimal trade-offs between performance and security in complex adversarial environments, ensuring UAV swarms can communicate effectively while maintaining operational security.

The breakthrough that enables rapid adaptation to changing environments is the Transfer Search-based Dynamic Constrained Multi-Objective Evolutionary Algorithm (TrS-DCMOEA). Unlike conventional optimization algorithms that restart from scratch with each environmental change, TrS-DCMOEA leverages historical knowledge through transfer learning. The algorithm maps previous optimal solutions into new environments through latent space representations, rapidly generating high-quality starting points for optimization in changed conditions.

The implications of this research are far-reaching. By enabling UAV swarms to make optimal resource allocation decisions under incomplete information and dynamic constraints, the framework removes a fundamental barrier to large-scale autonomous operations in contested environments. The validated approach provides a critically needed intelligent system for industries developing UAV communication networks, offering direct pathways to enhance both performance and security in uncertain conditions.

Looking ahead, Professor Lin's team plans to extend the framework to accommodate larger swarm configurations and integrate multi-dimensional resource management encompassing computation and storage alongside spectrum. Their goal is to provide robust technical support for developing highly efficient, secure, and adaptable UAV swarm systems, accelerating practical deployment across applications from low-altitude economy to integrated space-air-ground 6G networks.

Original Source

Kuixian Li, Jinjie Liu, Xin Gu, Yandie YANG, Cheng CHANG, Haipeng CHEN, Liangtian WAN, Yun LIN. Dynamic decision-making of UAV swarm based on constrained multi-objective optimization under incomplete interference information [J]. Chinese Journal of Aeronautics, https://doi.org/10.1016/j.cja.2025.103846.

About Chinese Journal of Aeronautics

Chinese Journal of Aeronautics (CJA) is an open access, peer-reviewed international journal covering all aspects of aerospace engineering, monthly published by Elsevier. The Journal reports the scientific and technological achievements and frontiers in aeronautic engineering and astronautic engineering, in both theory and practice. CJA is indexed in SCI (IF = 5.7, Q1), EI, IAA, AJ, CSA, Scopus.