Hybrid Framework for UAV Mission Planning Using Logic-Based Imaging and Reinforcement Learning

This study proposes an Unmanned Aerial Vehicle(UAV) mission planning framework that integrates Logic-Driven Imaging Planning (LDIP) with Reinforcement Learning-Based Path planning(RLPP). The proposed framework consists of the Mission-Specific Imaging Library (MSIL), which generates and stores mission-specific parameters, a module for calculating the Sweep Effective Flight Area (SEFA) based on Detectable Area(DA), the LDIP module to ensure image quality, and the RLPP module for trajectory optimization. In particular, the LDIP module generates imaging plans based on the given mission context, while the RLPP module optimizes the UAV path within the SEFA. Each module is designed to be adaptive to dynamic environments by considering AI performance, camera specifications, and UAV platform capabilities. The simulation results confirm that the proposed approach improves mission efficiency compared to conventional methods.