为解决强对抗场景下无人机因遭受干扰而导致GPS失能无法精确获取自身定位的问题，考虑到无人机经常以编队或集群形式行动，提出一种依靠编队内的无人机相互测量相对空间位置并互为定位的方法，使无人机在GPS信号丢失后依然可以实时更新自身位置。针对GPS拒止环境，引入部分可观测马尔可夫决策过程（POMDP）理论，分析了POMDP模型要素，建立起协同定位调度的POMDP决策模型。提出了基于扩展卡尔曼滤波（EKF）的信念状态更新方法和基于深度强化学习中深度Q网络（Deep Q-Network，DQN）的Q值估计方法，以实现协同实时精确定位。不同场景下的应用测试表明，所建立的模型能够实现编队中GPS正常无人机的高效管理调度，能够控制GPS正常无人机对GPS失效无人机进行有效协同定位，即模型有效性得到了验证。

In strong adversarial scenarios,?unmanned?aerial vehicles (UAVs) often experience GPS malfunction due to interference, making it difficult to obtain their positioning accurately. Since UAVs often operate in formations or clusters, this paper proposed a strategy that relies on drones within the formation to measure relative spatial positions and locate each other, allowing UAVs to update their position information in real time, even after GPS signal loss. Firstly, in response to the GPS-denied environment, this paper introduced the theory of the Partially Observable Markov Decision Process (POMDP), analyzed the model elements of POMDP, and establishes a collaborative positioning and scheduling POMDP decision model. Additionally, this paper proposed a belief state update method based on the Extended Kalman Filter (EKF), as well as a Q-value estimation method based on Deep Q-Network (DQN) in deep reinforcement learning to achieve accurate collaborative real-time positioning. Application tests in different scenarios have shown that the proposed model can achieve efficient management and scheduling of unmanned aerial vehicles in formation by effectively coordinating the positioning of GPS failure unmanned aerial vehicles with good control performance.