Abstract: The straw checkerboard is one of the most effective engineering sand fixations in Xinjiang of western China. Reed straw can often be utilized to produce the sand barriers of straw checkerboard, according to the regional conditions. Sand break and water conservation can be expected to serve as excellent ecological benefits. However, manual paving protocols of the reed straw checkerboard cannot fully meet the large-scale production at present, due to the high labor intensity and cost during straw-pressing operation. Alternatively, mechanical paving of the reed straw checkerboard can be expected to ensure the performance of its sand fixation. However, the existing research focuses mainly on the paving machinery in the longitudinal column of the straw checkerboard. By contrast, the transverse straw-pressing mechanism can be required for a certain spacing distance to press the straw. Among them, the straw is firstly tied in the shape of squares on the desert surface, and then interlaced by the equally spaced longitudinal and transverse columns. A grid of straw checkerboard is often formed to be generally laid first in the longitudinal columns in the same direction, as the forward direction of the machine, and then in the transverse columns. The existing transverse paving machinery usually combines the straw-pressing and straw-laying machine for instantaneous speed matching to realize the transverse column of the paving straw checkerboard. It is still lacking in the response relationship among the operating parameters of the transverse column straw-pressing mechanism on the quality of straw-pressing operation, as well as the motion and mechanical models. The paving quality of the straw checkerboard can dominate in the evaluation of the performance of a transverse straw-pressing mechanism. Taking the transverse paving of the reed straw checkerboard as the research object, this study aims to propose a “T” type straw checkerboard of the transverse and longitudinal synergistic paving mode. The transverse intermittent and longitudinal continuous straw-pressing were combined during paving. Two technical requirements were fully met in the paving mode. Specifically, the transverse straw-pressing was required to avoid the first longitudinal and the adjacent molded transverse reed sand barriers, as well as the intermittent zero-speed straw-pressing. The institutional kinematic model was constructed for the transverse straw-pressing, according to the manual action and motion trajectory. The agronomical requirements of the reed straw checkerboard were combined to determine the optimization objective of the transverse intermittent straw-pressing mechanism using the MATLAB and ADAMS platforms. The auxiliary optimization interface was written to optimize the parameters of the mechanism; The mechanical model was also constructed to explore the relationship between the straw-pressing cutter and the sand/reed straw. Particularly, the reed straw was often broken during pressing and extruding, indicating the large resistance in the stage of entering the sand. The structural parameters of the straw-pressing cutter were determined after optimization. The results show that there was relative consistency in the actual, ADAMS simulation and theoretical trajectory of the transverse intermittent straw-pressing mechanism. The qualified rates were 86.4% and 84.8%, respectively, for the transverse column depth and the edge thickness of the reed straw checkerboard. The relatively stable and reliable performance of transverse intermittent straw-pressing has verified the correctness of the theoretical design and the feasibility of the mechanism. The finding can provide a strong reference to optimize the transverse straw-pressing mechanism of the reed straw checkerboard paying machine.