Multi-UAV Joint Scheduling for Multimodal Data Sensing and Offloading in Post-Disaster Scenarios

In post-disaster scenarios where ground infrastructure is damaged, Unmanned Aerial Vehicles (UAVs) provide fast and flexible data offloading support. Unlike conventional UAV scheduling problems, post-disaster reconstruction requires UAVs to execute sensing-driven, heterogeneous, and delay-sensitive tasks in the absence of reliable ground communication and Internet of Things (IoT) support, thereby significantly increasing the complexity of task scheduling under limited onboard resources and multiple processing options. We propose a utility-driven scheduling framework that jointly considers task satisfaction and energy consumption fairness. A unified utility function is introduced, incorporating task type, delay constraints, data volume, and processing mode. This function is integrated into a coupled optimization model that captures the interdependence among task assignment, UAV path planning, and data processing. A hybrid Genetic-Assisted Local Optimization Algorithm (GALOA) is proposed to solve the Non-deterministic Polynomial-time hard (NP-hard) problem, which integrates genetic search and adaptive local refinement through parameter tuning and a perturbation mechanism. Through extensive simulation comparisons under varying problem scales, the proposed GALOA improves task delay by at least 7% and consistently outperforms baseline methods across different detection area configurations and repeated simulation runs, demonstrating its effectiveness and robustness in complex post-disaster scenarios.