Abstract: In large-scale Internet of Things scenarios, Wireless Sensor Network (WSN) often form multiple isolated sub-regions due to their spatial geographical distribution characteristics, and this geographical segmentation poses challenges to traditional data collection. Unmanned aerial vehicle (UAV), with their high mobility and flexible deployment, provide new insights into solving this problem. However, in existing UAV-assisted data collection schemes, UAV work independently and need to return to the base station separately to deliver data, leading to significant path redundancy and energy waste. To address this issue, a UAV Swarm Cluster-based Data Collection (USCDC) algorithm for wireless sensor networks was proposed. This algorithm adopts a two-layer UAV architecture: cooperative UAV are deployed at the bottom layer for data collection in each region, while sink UAV are configured at the top layer to uniformly transmit data to the base station. To achieve efficient operation of this architecture, a UAV Temporal Cooperative Scheduling (UTCS) algorithm was designed, which solves the coordination problem among UAV through a unified time coordinate system and event-triggered mechanism. Based on this, a regional path optimization method based on the Improved Ant Colony Optimization (IACO) algorithm and a sink UAV visiting sequence optimization strategy based on the Improved Particle Swarm Optimization (IPSO) algorithm were designed respectively. This hierarchical cooperation mechanism enables cooperative UAV to complete data transmission without returning to the base station, significantly reducing path overhead and energy consumption. Simulation results show that compared with existing data collection algorithms, the USCDC algorithm reduces path length and energy consumption by 52% and 35% respectively, verifying the effectiveness of this scheme in large-scale sensor networks data collection.