Simultaneous localization and mapping (SLAM) is paramount for unmanned systems to achieve self-localization and navigation. It is challenging to perform SLAM in large environments, due to sensor limitations, complexity of the environment, and computational resources. We propose a novel approach for localization and mapping of autonomous vehicles using radio fingerprints, for example WiFi (Wireless Fidelity) or LTE (Long Term Evolution) radio features, which are widely available in the existing infrastructure. In particular, we present two solutions to exploit the radio fingerprints for SLAM. In the first solution-namely Radio SLAM, the output is a radio fingerprint map generated using SLAM technique. In the second solution-namely Radio+LiDAR SLAM, we use radio fingerprint to assist conventional LiDAR-based SLAM to improve accuracy and speed, while generating the occupancy map. We demonstrate the effectiveness of our system in three different environments, namely outdoor, indoor building, and semi-indoor environment.

翻译：同时定位与地图构建（SLAM）对于无人系统实现自我定位与导航至关重要。在大型环境中执行SLAM因其传感器限制、环境复杂性及计算资源等问题而具有挑战性。我们提出了一种利用无线电指纹（例如现有基础设施中广泛存在的WiFi（无线保真）或LTE（长期演进）无线电特征）对自主车辆进行定位与地图构建的新方法。具体而言，我们提出了两种利用无线电指纹进行SLAM的解决方案。第一种方案——即无线电SLAM，其输出为采用SLAM技术生成的无线电指纹地图。第二种方案——即无线电+激光雷达SLAM，我们利用无线电指纹辅助传统的基于激光雷达的SLAM，在生成占用地图的同时提高精度与速度。我们在三种不同环境（即室外、室内建筑及半室内环境）中验证了系统的有效性。