This paper studies low-altitude mission test design for UAV swarm ground missions in complex urban environments. Traditional test design workflows depend heavily on expert-crafted rules and static settings, which limits adaptability under dynamic mission conditions. To address this issue, we propose an intelligent framework that combines a Multi-Stage Constrained Multi-Objective Optimization algorithm with Proximal Policy Optimization-based adaptive hyperparameter tuning. The framework optimizes resource allocation by balancing mission effectiveness, mission risk, and mission cost under mission constraints. Simulation results show improved convergence behavior, solution quality, and robustness compared with baseline settings.
Miao et al. (Tue,) studied this question.