传统的基于两两检测的冲突检测方法效率较低，为满足大规模场景下高可靠性、高性能的要求，本文提出了多级空间哈希冲突检测方法。该方法基于动态空间划分以及航线分割，使用四层空间哈希表，能够快速定位潜在冲突区域。与以往基于离散点检测的方法不同，本方法采用连续模型距离检测，准确性更高。

   针对检测到的冲突，本文提出基于优先级序列的解脱方法。该方法在不改变计划路径和速率的情况下，按照优先级依次调整航空器到达各关键点的时间来避免飞行冲突。与在检测与解脱间迭代的传统模式相比，该算法仅需一遍就能给出正确的调整方案。仿真结果表明，本文设计的算法在大规模场景下能够更加高效地发现飞行冲突，并快速解脱。   此外，本文还提出了基于A*算法搜索合理时间序列的最优化解脱方法。实验示例表明利用A*算法的启发函数，并考虑冲突数量和要调整的航空器数量，该算法能非常有效地指导结点选择，快速到达总调整时间最小的目标最优解。

   最后，应用上述算法，本文设计并实现了基于C/S架构的三维空域管理系统。该系统可作为空域管理辅助决策工具，帮助用户管理和控制各类飞行事务，有效防止空中交通事故的发生。

     With development of aeronautical technologies, and changes of international situation, air traffic worldwide is becoming more complicated and busier, thus airspace management is confronted with a bigger challenge now, especially for military applications involving many kinds of aircraft. Therefore, conflict detection and resolution in large scale scenarios is one of key technologies for national security and development of future air traffic management.

     Traditional conflict detection method measuring on all aircraft pairs is inefficient. To meet demands for high accuracy and performance in large scale scenarios, we devise a multi-level spatial hashing approach for conflict detection. By dynamic space and trajectory division, the approach can locate potential conflicts quickly with four-layer spatial hash table. Different from discrete-point-based methods, we execute detections on continuous segments to be more precise.

    For conflicts detected, we propose a priority sequence based resolution method. According to priority, the method sequentially regulates when aircraft arrives at each key point, without changing planned path and velocity. Compared with traditional iterative detection-resolution model, our method just needs one pass to make correct adjustment. The results of simulation show that our algorithms can locate conflicts more efficiently and solve them more quickly in large scale scenarios.

     Besides, we also present an A* based algorithm to search optimal resolution with reasonable time sequence. Experimental examples show that, with the using of the heuristic function of A* method and considering of the number of conflicts and adjusted aircraft, this algorithm guides node choosing very efficiently, and steps to the optimal solution quickly.

At last, applied above methods, a three-dimensional airspace management system on C/S architecture is designed and implemented. As a decision support tool for airspace management, this system can help users manage and control kinds of flight missions, and avoid flight accidents efficiently.