Abstract: The autonomous navigation of mobile robots in complex environments relied heavily on environmental maps， the accuracy of map construction depended on system optimization algorithms and the performance of supporting sensors， the path planning also needed to coordinate two core links：global planning and local planning. Therefore， based on the theory of Simultaneous Localization and Mapping （SLAM）， an experimental platform for mobile robots was built， an autonomous navigation method based on laser SLAM was proposed， and an optimization strategy combining time optimality and graph optimization was designed for the local planning module in path planning. To verify the system performance， the simulation experiments and physical experiments were conducted， and the key indicators such as the calibration accuracy of robot sensors， the accuracy of environmental map construction， and the rationality of local path planning were tested with emphasis. The results showed that the designed mobile robot could accurately complete the tasks of map construction and path planning， which provided a theoretical basis for the subsequent application of robots in forest operation scenarios.