Abstract
The development of research of unicycle Zero Torque Balance Controller (ZTC) in our lab has been already mature. ZTC can estimate the position of center of gravity of unicycle making the rider riding not interfere the balancing. This thesis inherits this basis and combines two unicycle to a parallel unicycle trying to solve the significant mass variation circumstance that the balance controller has to balance in both single rider and two riders situation. First, we derive the physics model of the parallel unicycle for simulation. Then we use inverted pendulum dynamic equation to approximate the original model and linearize in operation point make it more easily for designing controller. Then we consider the mass variation situation in Lyapunov function, transform it into LMI form and solve the ZTC parameters through Matlab LMI Toolbox that satisfy the condition of system stability and performance. Finally, according to the model, dynamic equation and simulation, we successfully implement an automatic balanced, electric assisted, pedaled, parallel unicycle, inverted-pendulum-based vehicle include mechanism, signal processing, chip programming and motor control.