Observation optimization for marine target tracking by airborne and maritime unmanned platforms cooperation using adaptive enhanced dung beetle optimization.

Cooperative observation optimization for maritime targets is crucial for improving marine monitoring precision. Existing research predominantly focuses on homogeneous platform cooperative observation optimization under random error influences, while neglecting the collaborative optimization challenges of heterogeneous platforms affected by systematic errors. To address these limitations, this paper proposes a heterogeneous unmanned platform cooperative observation optimization method based on Adaptive Enhanced Dung Beetle Optimization (AEDBO) algorithm. First, we derive optimal observation configurations for the aerial unmanned platform, maritime cooperative unmanned platform, and targets under azimuth systematic error impacts using an attitude correction algorithm. Subsequently, we design AEDBO by integrating an improved Tent chaotic mapping and centroid opposition-based learning strategy to enhance population diversity. An adaptive convergence factor and nonlinear ball-rolling dung beetle population decline model are introduced to balance global exploration and local exploitation capabilities, while Cauchy-Gaussian mutation strategies are employed to prevent premature convergence. Finally, AEDBO is applied to aerial unmanned platform trajectory optimization. Experimental results demonstrate that compared with DBO, AEDBO achieves improved optimization accuracy across 15 benchmark functions. In both safe and hazardous zone scenarios, optimized trajectories reduce target tracking errors to below 10 m, with optimal observation configurations validated through practical experiments. This study establishes a novel theoretical framework and optimization toolkit for heterogeneous unmanned platform cooperative observation.