To address the problems of multiple aircraft coordination constraints， frequent conflict risks， and operational inefficiencies faced by UAV cluster flights in complex urban environments， a collaborative deduction and optimal allocation method for urban low-altitude UAV flight plan is proposed. Firstly， for the UAV “individual”， the conflict between UAVs and urban static environment targets is considered， and a three-dimensional path planning method based on probabilistic risk maps is constructed to achieve the initial generation of UAV flight plan. Then， for the UAV “group”， considering the conflict between UAVs， a multi-type conflict cooperative deduction model for UAV flight plan is constructed， and the conflict characteristic indicators such as conflict occurrence rate， conflict risk level， and grid occupancy ratio are designed and metricized. Finally， diversified allocation strategies such as adjusting flight paths， flight speeds， and flight times are designed. Based on the multi-strategy self-adaptive configuration， an optimization and allocation model is established. The optimization performance and parameter sensitivity of the proposed method are verified， demonstrating the optimization performance and parameter sensitivity of urban low-altitude flight plans under different traffic density scenarios. Experiments show that the proposed method can reduce flight plan conflicts by 96.2% on the basis of effectively controlling the risk cost and time cost， completely conflict free flight plan can be obtained for UAV operation scenarios with less than 40 UAVs， and the minimum flight conflict resolution rate for UAVs with less than 100 UAVs can be controlled above 95%. The proposed method is scientific and effective， and can provide a theoretical basis and methodological guidance for safe and efficient management of low-altitude UAV flight activities in complex cities.