To address load–energy dynamic coupling in heterogeneous unmanned aerial vehicle (UAV) emergency rescue, this paper proposes an energy-coupled heterogeneous UAV task allocation (EC-HUTA) model that explicitly characterizes nonlinear interdependencies among payload, velocity, and power consumption, minimizing aggregate mission costs subject to physical and temporal constraints. To tackle the resulting high-dimensional, nonconvex problem, we introduce a multi-strategy improved stellar oscillation optimizer (MISOO), establishing a closed-loop synergistic system through three coupled stages: (i) evolutionary game-theoretic strategy competition via replicator dynamics for adaptive exploration–exploitation balance; (ii) intuitionistic fuzzy entropy (IFE)-driven dimension-wise parameter control, where IFE calibrates global exploration intensity while dimension-specific crossover probabilities accommodate heterogeneous convergence; and (iii) memory-driven differential escape mechanisms modulated by historical memory parameters to evade local optima. Cross-stage coupling through IFE ensures state information flows across the “strategy selection-refined search-dynamic escape” pipeline. Coupled with a dual-layer encoding scheme, this framework ensures efficient feasible search. Ablation studies validate each mechanism’s contribution. Evaluations on CEC2017 benchmarks demonstrate MISOO’s superior convergence against six metaheuristics. Large-scale earthquake rescue simulations confirm that EC-HUTA/MISOO strictly adheres to nonlinear energy constraints while enhancing task completion and temporal compliance. These results validate the framework’s efficacy for time-critical emergency resource allocation.
1 September 2025 in Tsinghua Science and Technology
Due to the rising demand for Quality of Service (QoS) in the emerging network, this paper considers the routing problem of efficiently planning the transmission path of flow demands to balance the load of the network. Existing classical routing algorithms based on static decision-making are unable to adjust to dynamical scenarios with fluctuating flow. Traditional reinforcement learning-based methods face challenges in large-scale scenarios due to the vast state and action spaces. Besides, the neural networks used for policy generation also suffer from issues like rudimentary feature extraction and limited learning capability. In this paper, we first introduce RequestNet, a novel attention-based model designed to better leverage the information of network topology and feature correlations. We also propose a multi-agent modeling approach for the network packet routing problem, named RequestNet-MA, which treats pairs of edge routers as agents, allowing us to allocate traffic from the perspective of flow demands, thus significantly reducing the dimensions of state and action spaces and enhancing inter-agent communication and cooperation. Extensive experiments and ablation studies demonstrate that the proposed RequestNet-MA can reduce maximum link utilization and scale well to a large-scale network.