On position/force tracking control problem of cooperative robot manipulators using adaptive fuzzy backstepping approach

摘要

Abstract In this paper, the position and force tracking control problem of cooperative robot manipulator system handling a common rigid object with unknown dynamical models and unknown external disturbances is investigated. The universal approximation properties of fuzzy logic systems are employed to estimate the unknown system dynamics. On the other hand, by defining new state variables based on the integral and differential of position and orientation errors of the grasped object, the error system of coordinated robot manipulators is constructed. Subsequently by defining the appropriate change of coordinates and using the backstepping design strategy, an adaptive fuzzy backstepping position tracking control scheme is proposed for multi-robot manipulator systems. By utilizing the properties of internal forces, extra terms are also added to the control signals to consider the force tracking problem. Moreover, it is shown that the proposed adaptive fuzzy backstepping position/force control approach ensures all the signals of the closed loop system uniformly ultimately bounded and tracking errors of both positions and forces can converge to small desired values by proper selection of the design parameters. Finally, the theoretic achievements are tested on the two three-link planar robot manipulators cooperatively handling a common object to illustrate the effectiveness of the proposed approach. Highlights ? An adaptive backstepping method is proposed for the tracking problem of cooperative manipulators. ? Cooperative manipulators with unknown dynamics and external disturbances is studied. ? The fuzzy logic system is applied as an approximator. ? The error system is constructed by the integral and differential of position and orientation errors of the grasped object. ? It is proved that the closed loop systems are uniformly ultimately bounded. ]]>