An Autonomous Balancing Control for a Two-Legged Wheeled Robot

This paper presents an advanced linear quadratic control for a two-legged wheeled robot (TLWR). The TLWR model with an adjustable height mechanism is introduced to point out the challenges of this kind of robot. In order to guarantee the balancing of the TLWR, three linear quadratic controls are designed according to the three different poses, which are low, medium, and high levels. A supervisor is used to select the suitable control for the TLWR according to the robotic height. In order to evaluate the effectiveness of the proposed method, a real TLWR test bench is constructed with a graphic user interface, which is built to acquire and set the controller and robot online through wireless communication. In practice, the robotic height is indirectly estimated from the hip angle, and the pitch angle of the robot is modified according to the robotic posture. Two experiments are conducted on the practical test bench to demonstrate the superiority of the proposed control.