The structured and networked planning of urban low-altitude air routes is an effective approach to addressing constraints on low-altitude airspace resources and enhancing operational safety. The proposed methodology is a combined clustering-based "trunk-feeder" route network planning method for urban low-altitude logistics scenarios. The incorporation of takeoff and landing site layouts, in conjunction with actual operational traffic data, serves to address the practical demands of route and traffic considerations. Initially, optimal trunk route directions are determined using either principal component analysis (PCA) or numerical optimization. The quality of clustering is evaluated through the use of PCA-based or geometric similarity models, yielding four different approaches to the clustering of route networks. Secondly, a comprehensive evaluation system encompassing the total route network length, construction costs, and route intersection points is established to assess the performance of the four clustering methods. Finally, the feasibility and effectiveness of the method were validated using Shaoxing City as a case study. In comparison with the initial scenario, the high-density scenario resulted in a 21.39% reduction in total network construction costs and a 34.4% decrease in the number of route intersections. In the medium-density scenario, total network construction costs decreased by 36.4%, and the number of route intersections decreased by 16.9%. The findings indicate that the proposed "numerical optimization-geometric similarity" combined clustering method significantly enhances low-altitude operational safety and economic efficiency, thereby providing theoretical foundations and technical support for the scientific planning of urban low-altitude traffic systems.