Abstract: The layered collaborative technology of low-altitude logistics of unmanned aerial vehicles (UAVs) serves as the core support for breaking through bottlenecks in full-process logistics distribution. Centering on the three-layer architecture of collaborative task allocation, collaborative trajectory planning, and dynamic trajectory replanning, this paper provides a comprehensive overview of the research progress in key technologies. The collaborative task allocation layer focuses on order clustering and multi-UAV matching, covering algorithms based on optimization models (exact, heuristic, and meta-heuristic optimization), market mechanisms (auction bidding and game-theory-based alliance optimization), and swarm intelligence (swarm intelligence optimization and reinforcement learning). These algorithms are specifically designed to address the challenges of multiconstraint coupling and collaborative task allocation. By adopting techniques such as constraint decoupling, collision-free trajectory design, and priority-aware algorithms, this layer improves the allocation efficiency and dynamic adaptability of the system to ensure that UAVs are assigned to the corresponding logistical tasks in complex and changing environments. The collaborative trajectory planning layer aims to generate three-dimensional airspace paths by combining algorithms for precise optimization (accurate optimization and multiconstraint planning), swarm intelligence (evolutionary optimization and swarm optimization), and intelligent fusion (reinforcement learning-based planning and hierarchical hybrid-driven planning). By integrating these algorithms, this layer achieves a balanced consideration of three critical factors: obstacle avoidance, energy consumption, and operational efficiency. This balance not only enhances the safety of UAV trajectories but also improves the generalization ability of the trajectory planning system to enable it to adapt to different low-altitude logistics scenarios (e.g., urban residential areas, suburban industrial parks, and rural areas with varying geographical features). The dynamic trajectory replanning layer targets abrupt environmental changes such as sudden weather disturbances (e.g., strong winds and sudden rain), temporary airspace restrictions, and unexpected obstacles during UAV flight. Relying on search and traversal algorithms (deterministic graph search and random hierarchical search), physical modeling (artificial potential field-based local planning and heuristic-assisted physical replanning), and intelligence and learning (intelligent optimization, reinforcement learning and deep adaptive planning), this layer realizes rapid adjustments to UAV trajectories. Such adjustments are crucial to ensure the real-time performance of the logistics system (allowing UAVs to respond promptly to unexpected situations without significant distribution delays) and safeguarding system robustness (ensuring stable logistics operations amid environmental uncertainties). A comprehensive analysis shows current technical bottlenecks: strong multiconstraint coupling (interactions between flight time, load capacity, airspace regulations, and energy limits hinder optimal task allocation and trajectory planning); insufficient dynamic adaptability (existing algorithms struggle to adjust quickly to logistics environment changes or sudden task requirement shifts); low large-scale collaboration efficiency (complex coordination among multiple UAVs reduces the overall operational efficiency); and scenario disconnection (many research results based on idealized simulations fail to align with actual complex low-altitude logistics scenarios, causing application difficulties). Looking ahead, future innovations should move toward cross-layer collaborative optimization (integrating the three layers for overall system optimization instead of independent layer optimization); scenario-specific customization and integration (developing targeted technologies/algorithms based on practical scenario characteristics to enhance adaptability and practicality); large-scale swarm intelligence (further exploring swarm intelligence to improve coordination efficiency of large UAV clusters); dynamic robust design (strengthening research to maintain system stability amid complex unexpected environmental changes); and green energy conservation (developing energy-saving UAV technologies and optimizing trajectory planning to reduce energy consumption, in line with green sustainable development). Advancing these innovations will promote the large-scale implementation of low-altitude logistics UAV systems to transform and upgrade the logistics distribution industry.