Abstract: To address the challenge of logistics UAVs navigating through urban obstacles under time constraints, a hierarchical trajectory planning method was developed. First, a real residential area map was transformed into a 2D grid map, with grid risk assessed by the density of surrounding obstacles. The planning process was split into two stages: front-end path planning and back-end trajectory optimization. In the front-end stage, key points were extracted from the path points generated by the A* algorithm, followed by linear interpolation to form a preliminary path. In the back-end stage, the trajectory was optimized by minimizing snap (the second derivative of acceleration), and a time allocation method was designed to meet time constraints. Simulation experiments demonstrated that this method not only ensured flight safety under tight time constraints but also achieved lower energy consumption compared to other methods. This hierarchical approach provides an effective solution for urban low-altitude logistics UAV trajectory planning, balancing time constraints, risk, and energy efficiency.