Fuzzy logic self-motion planning and robust adaptive control for tip-over avoidance of redundant mobile modular manipulators

A redundant nonholonomic mobile modular manipulator composed of a 3-wheeled nonholonomic mobile platform and a N-degree of freedom (DOF) onboard modular manipulator is investigated in this paper. Redundancy of the atop manipulator is used to avoid tipping over of the entire robot through automatically adjusting self-motions in a realtime manner. On the basis of modular robot concept, a kind of integrated dynamic modeling method is proposed. A realtime fuzzy logic (FL) self-motion planner and a robust adaptive controller are presented to prevent die robot from overturn without affecting the end-effector specified tasks. A new sliding mode function is devised, which is not only continuous but also infinitely differentiable, then chattering caused by classical sign function can be eliminated effectively. The proposed algorithm does not need exact a priori dynamic parameters and has strong external disturbance suppression ability. Simulation results for a real robot moving on a slope demonstrate that the proposed controller is effective.