Facing complex non-structural environments, such as post-disaster rescue, pipeline inspection, industrial maintenance, and scientific investigation, etc., higher requirements are placed on the robot’s ability to pass through, but the existing robots are not well adapted to complex environments. In this paper, based on the flexible spinal structure of animals, we combine the characteristics of tracked robots and reconfigurable robots to design mobile robots with both high obstacle-crossing and flexibility. The maximum height of the robot and its constraints are analyzed from the kinematic point of view, and the dynamics model of the robot’s obstacle-crossing process is established by using the Newton-Euler method, and the driving torque is derived according to the maximum height of the obstacle-crossing and its constraints. According to the maximum obstacle-crossing height derived from the theoretical calculation, a robot simulation platform was established to simulate the obstacle-crossing process of the robot and verify the correctness of the theoretical calculation.