Arduino is used as the core controller and 2D lidar sensor is used as the external environment detector in the construction of mobile robot SLAM.The collected signal is processed by Arduino and transmitted to ROS operating system through WiFi. The environment map is constructed through processing by the upper computer.The traditional RBPF-SLAM algorithm was optimized, and the map was constructed by the mobile robot test platform under the ROS robot operating system in two different environments. The composition effects before and after the algorithm optimization were compared to compare the performance of the two algorithms.In this paper, one of the experimental environments was measured. By comparing the scale accuracy of the map constructed by the two algorithms, the performance of the algorithm before and after optimization was quantified.It can be seen that the optimized algorithm has better effect of 2D plane map construction through the demonstration of two methods of visual map construction effect and map construction accuracy.