2019, 40(3): 302-310.
doi: 10.21656/1000-0887.390163
Abstract:
The control of underactuated systems was an important field of nonlinear control. The underactuated system refers to a nonlinear system with an input control variable number less than the degree of freedom number. At present, the main methods of dynamics and control research of underactuated nonlinear systems include the linear quadraticform optimal control method and the partial feedback linearization method, and how to make the system stabilize in the equilibrium position is always a difficult point. For the virtual constraint method, the periodic motion of the system is designed with a selected cyclic independent variable. Based on the typical underactuated model crane, the virtual constraint method was adopted to make the system stabilize or oscillate in equilibrium position. First, through establishment of the virtual constraints, the system’s degrees of freedom were reduced. Then, the system state equations were derived according to the partial feedback linearization theory. Finally, the feedback controller was designed with the linear quadraticform regulator. The simulation results show that, the weight can reach a stable state near the vertical position under the feedback control, which reflects the effectiveness of the virtual constraint method for underactuated systems.