面向无人平台的室内精密定位与构图技术

针对无人作战平台在室内等复杂环境下的高精度定位与地图构建问题，提出了一种利用信标锚节点辅助的单目视觉同步定位与地图构建方法，该方法通过WLAN锚节点估计出单目视觉尺度因子并结合视觉传感参数，获取真实尺度空间下的定位位置，并利用因子图模型对WLAN定位和单目视觉定位结果进行最大后验概率准则下数据融合，从而获得精确的位置和构图信息。实验表明：文中所提出的方法较好地解决了传统单目视觉定位中的尺度不确定问题，定位精度达到分米级，扩展了单目视觉SLAM在真实尺度空间中的应用，在复杂环境下的无人侦察、搜救和打击等应用领域具有重要的工程和实践意义。

Indoor Precision Positioning and Mapping Technology for Unmanned Platforms

In order to realize the hgh precision positioning and map construction for unmanned combat platforms in complex environments, we proposed a precision positioning and mapping method based on monocular visual SLAM with WLAN adided. This method can estimate the scale factor of monocular VSLAM and get the position in real scale space. Furthermore, the result of the WLAN location and the monocular vision positioning is fused by the factor graph model, realizing accurate positioning and sparse mapping. The experiment shows that this method can estimate scale factor well and reach the positioning accuracy of decimeter, and has important engineering and practical significance in applications such as unmanned reconnaissance, search and resue in complex environments.