Under modern conditions, the development momentum of international unmanned driving technology is booming, but the industrialization of unmanned driving cars is still limited by the current level of software and hardware. Reducing the complex and high standard requirements of unmanned driving and then studying unmanned vehicle navigation technology in some simple scenarios are also a hot topic in current research. This paper studies an unmanned autonomous navigation vehicle based on robot operating system (ROS) software framework on the Ubuntu operating system. It combines algorithms such as simultaneous localization and mapping (SLAM) technology, adaptive Monte Carlo localization (AMCL) technology and path planning, aiming to solve the problems of simultaneous localization, mapping and path planning in a closed environment. This case algorithm avoids the problem of large error accumulation caused by traditional odometer navigation and global positioning system (GPS) navigation technologies. The simulation results show that the algorithm has good practical application value.