To address the limitations of traditional unmanned aerial vehicle （UAV） path planning algorithms， which often exhibit low efficiency and low success rate of path planning due to inadequate consideration of environmental threats such as terrain elevation， radar detection， and anti-aircraft fire in adversarial scenarios， this paper proposes an improved ARA*-based path planning algorithm for UAVs. The algorithm achieves the integrated calculation of threat factors such as terrain elevation， radar detection， and anti-aircraft fire through the variable weight theory， generates a global 3D gridded threat map， and conducts path planning with an improved ARA* algorithm on the threat map. It incorporates UAV flight performance in path planning constraints for realistic trajectory planning. Simulation experiments demonstrate that the proposed method exhibits excellent adaptability and effectiveness in complex adversarial scenarios， significantly reducing the threat level of planned paths while maintaining a high success rate compared to traditional algorithms. This approach offers a promising solution for enhancing UAV mission completion rates in high-threat environments.