To address the challenges of poor search efficiency

non-fluid pathways

and inadequate obstacle evasion in conventional A* algorithms

an innovative globally-guided dual-layer planning algorithm known as G-DWA* is proposed in this study

which combines the A* algorithm with the dynamic window approach (DWA). Initially

a model for obstacle coverage is formulated

the weight coefficient for the A* heuristic function is adjusted dynamically

and a line-of-sight algorithm is implemented to extract pivotal path points. Then

a discount factor is employed to enhance the original sub-function evaluation of DWA

introducing novel assessment metrics for velocity smoothness and trajectory likeness. Finally

the essential path nodes obtained from the refined A* algorithm serve as intermediate waypoints for the augmented DWA. The impact radius of each waypoint is computed

providing comprehensive guidance for localized planning. Experimental results indicate that the G-DWA* algorithm considerably improves the velocity and fluidity of autonomous path planning for drones. It is proficient in evading unidentified obstacles based on global optimization

thereby effectively fulfilling the requirements of intricate planning tasks

inclusive of military operations.