Hierarchical Encoding-Decoding for 3D Fixed-Wing UAV Formation Optimization Using an Improved Exponential-Trigonometric Algorithm

A novel Exponential-Trigonometric Optimization (ETO) algorithm integrating chaotic mapping and directional search strategies, is proposed in this paper for fixed-wing unmanned aerial vehicles (UAV) swarm formation optimization. Traditional methods, limited to two-dimensional planar formations, inadequately meet the demands of modern three-dimensional UAV formation combat scenarios that require dynamic spatial coordination. To bridge this limitation, a comprehensive optimization framework is established, incorporating a hierarchical three-dimensional UAV formation encoding-decoding mechanism. This mechanism decomposes complex spatial configurations into layered optimizable parameters, enabling more precise control over UAV swarm formations. Furthermore, the improved ETO (IETO) algorithm employs circle chaotic mapping to dynamically optimize the initial population distribution and implements a normal cloud model-based directional search strategy to eliminate poorly adapted individuals within the swarm. Through this hybrid approach, the proposed IETO achieves superior convergence speed and enhanced optimization precision compared to conventional algorithms. Extensive simulations validate the algorithm’s superiority, demonstrating that the proposed algorithm efficiently addresses swarm formation optimization with higher accuracy and faster convergence than existing algorithms.