In disaster-affected areas, effective UAV path planning is essential for precise data collection, prompt resource delivery, and safe navigation through challenging terrain. This work proposes a hybrid path planning methodology for such scenarios by dividing the disaster zone into open areas for global path planning and obstacle-dense zones for local path planning. A representation based on a hexagonal grid is utilized for local planning, which reduces computational redundancy and offers efficient waypoint selection. The combination of dynamic obstacle recognition and avoidance using LiDAR sensors and the A* algorithm enables real-time adaptation to changing situations. A spline smoothing technique is used to further improve the resulting paths.