Path planning is critical for autonomous drone flight. Considering the high computational demands of traditional graph search algorithms for path planning under multi-drone conditions, this study proposes an improved flow-field pathfinding optimization algorithm extended to three-dimensional space. This approach better enables real-time 3D path planning for multiple drones operating under flight constraints. The algorithm addresses the excessive redundant planning costs inherent in traditional graph search algorithms when handling multi-objective path planning problems. Simulation results demonstrate that the new algorithm effectively overcomes the limitations of traditional graph search methods, successfully solving path planning problems from multiple starting points to a single target. It holds significant application value for three-dimensional online path planning in UAVs.