The paper presents the solution and algorithm for balancing control for human transport robots when moving up and down stairs. The solution for balancing human transport robots through balancing control of the human transport mechanism is the robot’s seat. The balancing control process when the robot moves up and down stairs is performed through controlling the linear actuator combined with sensor signals that determine the robot’s tilt angle. The dynamic model of the robot balancing control system is built based on the solution for balancing with an electric cylinder. The sliding control algorithm with PID sliding surface is built to control the balancing mechanism for the robot. In order to increase the efficiency of the sliding controller, the Sugeno fuzzy controller is used to help calibrate the parameters for the controller. The control algorithm is compared with the basic SMC (Sliding Model Control) algorithm and the PID (Proportional Integral Derivative) controller. The simulation is performed on Matlab Simulink software. The experimental results show the effectiveness of the controller and the constructed solution, and that the FSMC-PID (Fuzzy Sliding Model Control-Proportional Integral Derivative) controller is better than the conventional SMC controller and the PID controller, and the chattering phenomenon in the sliding controller is overcome. The controller response still ensures to follow the reference value when the load and slope change, which shows the robustness of the controller for the robot balancing process.