针对无人机在城市空中交通（Urban Air Mobility, UAM）环境中面临的卫星定位不稳定和信号干扰问题，以及现有额外定位系统对机载算力需求较大的挑战，研究提出了一种轻量化的无人机视觉定位系统。通过开发融合状态空间模块的图像特征提取框架，显著提高了预测精度，实现了基于高斯金字塔的三元组自监督训练方法，以增强算法的鲁棒性。通过引入基于滑动窗口和相似度矩阵的特征匹配策略，优化了特征匹配过程并显著提高了推理速度。通过在Airsim仿真平台及真实UAM飞行场景中的实验验证，该算法在复杂环境中能够提供精准的额外定位数据。在多组消融实验和性能测试中，验证了算法的先进性与实时性，并有效降低了算力需求。结果表明，该系统在提高无人机定位精度的同时，为城市空中交通环境下的无人机定位提供了可行的视觉解决方案。

In view of the problems of unstable satellite navigation and positioning and signal interference faced by drones in the Urban Air Mobility (UAM) environment, as well as the challenges of existing additional navigation systems requiring greater airborne computing power, the study proposed a A lightweight UAV visual navigation and positioning system. By developing an image feature extraction framework that fuses state space modules, the prediction accuracy is significantly im-proved, and a triplet self-supervised training method based on Gaussian pyramid is implemented to enhance the robustness of the algorithm. By introducing a feature matching strategy based on sliding windows and similarity matrices, the feature matching process is optimized and the inference speed is significantly improved. Through experimental verification on the Airsim simulation platform and real UAM flight scenarios, this algorithm can provide accurate additional navigation and positioning data in complex environments. In multiple sets of ablation experiments and performance tests, the advanced nature and real-time performance of the algorithm were verified, and the computing power requirements were effectively reduced. The results show that this system not only improves the accuracy of UAV navigation, but also pro-vides a feasible visual solution for positioning and navigation in urban air traffic environments.