This article mainly aims at the problem of low accuracy of environmental mapping of wheeled robot system. Based on the theoretical and analytical research of wheeled robot kinematics model, sensor model and fusion algorithm, the model of wheeled robot system is established. Finally, through the simulation experiment, the impact of the performance parameters of laser radar (field of view, angular resolution) on the position and pose of wheeled robot and the quality of mapping during the environmental mapping process is comprehensively analyzed. It is found that the position uncertainty curve in the direction of the x and y coordinate axes shows a periodic downward trend with time, and the angle error shows a trend of oscillation up and down around a value. As the angle of view of the lidar increases, the curve oscillation decreases and tends to be smoother, and the white boundary of the mapping results becomes clearer and more complete. As the angular resolution of the lidar increases, the curve oscillations increase, and the amplitude is larger. At the same time, the white boundary of the mapping results is more blurred, and the accuracy is lower. Therefore, the accuracy of robot system environment mapping can be improved by increasing the angle of view of the lidar and reducing the angular resolution.