Energy-Efficient 3D Drone Navigation using A* Algorithm

Autonomous drones are powering the revolution in delivery,monitoring,rescue, and surveillance operations, where the real-time navigation of complex 3D environment is crucial. Nevertheless, conventional methods are usually required to choose between power efficiency and adaptability to changing conditions. A hybrid reinforcement learning and A*-based energy-efficient motion planning system with local dynamic obstacle integration is presented in this paper. The energy-aware heuristic is obtained from a physics-based engineering model that considers both horizontal and vertical components of the motion energy, thus making sure that optimization is carried out in a fair manner.Voxelized 3D Grid along with a LiDAR-like sensing mechanism facilitates Dynamic Obstacle Perception and Local Path Re-planning. The innovative adaptive architecture changes the ratio of energy and distance costs based on the battery condition, hence increasing durability and safety. Simulations indicate that this hybrid will be A*+RL approach achieves 92.4% success, 65% fewer collisions and 24% improved energy efficiency compared to conventional A*. The system has real-time adaptability and scalability for Larger environments and considerable enhancement in energy- Aware UAV navigation.