中文摘要：

对移动机器人未知环境中自主导航和SLAM（即时定位与构图）问题进行讨论。设计了一种构建2D可视化路标特征地图的方案,该方案结合单目视觉传感器和里程计的鲁棒感知模型,建立包含世界坐标系下三维信息的路标数据库,并获得全局环境下特征地图;提出了一种基于Python平台分析移动机器人自主导航鲁棒性的方法,通过在Python平台下引入扩展模块NumPy、PyLab构建仿真平台,对不同噪声环境下EKF-SLAM（扩展卡尔曼滤波器）和FastSLAM算法的导航过程进行了研究。实验显示了构图方案和仿真平台的可行性。

英文摘要：

Mobile robot autonomous navigation and SLAM（Simultaneous Localization and Mapping） in an unknown environment are discussed.A method to build two-dimensional visual landmark feature maps is designed,and based on monocular vision sensor and odometer of the robust perceptual model,a three-dimensional information landmark database in the world coordinate system is built and feature map is gained.A method to analyze the autonomous navigation of a mobile robot based on Python by means of putting into extended module NumPy and PyLab is put forward.On Python a simulation platform is constructed,and the navigation process is analyzed by EKF（Extended Kalman Filter）-SLAM and FastSLAM algorithm under different noise environments.Results show that the composition program and simulation platform are suitable.