Currently, drone path planning faces problems of high cost, low efficiency, and insufficient safety, which limits its application in complex urban environments. To overcome these challenges, this article proposes an improved 3D path planning algorithm based on the A * algorithm. Let's first simulate the flight environment using the upgraded grid method to ensure a safe flight area, and then add a calculation of safety costs to make flight route planning safer. Next, based on the different stages of flight route planning, flexibly adjust the weights and design a calculation method for flight costs, which can make the flight route shorter and more efficient in planning. Finally, by expanding the line of sight strategy and reducing redundant checks in path search, the A * algorithm solves the problems of numerous nodes, high computational complexity, and low efficiency in path planning. After in-depth and detailed experimental verification, the improved A * algorithm has achieved significant optimization in the time efficiency, number of nodes, and flight cost of path planning, and the planned route is smoother and more practical. The performance of this algorithm has been fully tested in complex urban environments, providing strong support for research on drone path planning in urban environments.