Symmetrical adaptive variable admittance control for position/force tracking of dual-arm cooperative manipulators with unknown trajectory deviations

摘要

In this paper, a new symmetrical adaptive variable admittance control is proposed for position/force tracking of dual-arm cooperative manipulators. Thanks to this control approach, the dual-arm cooperative manipulator is able to track a dynamic desired force and compensate for the unknown trajectory deviations, resulted from in terms of external disturbances and calibration errors. An object-oriented motion planning approach for dual-arm cooperative manipulators is adopted firstly, the motion of the single arm in the system is computed via closed-chain constrains. The trajectory deviations from external disturbance and calibration errors are considered, which will lead existing unknown external forces and internal forces to the closed-chain system. In particular, the modeling of closed-chain which including dual-arm and object is analyzed. According to the principle of "nonsqueezing" pseudoinverse, the forces acting on the center of the object are decomposed into external and internal forces. Furthermore, the external and internal forces are transformed to the tip of the object by load distribution strategy. To track the desired position and forces, a symmetrical adaptive variable admittance control for dual-arm coordination is achieved for the first time by adjusting the admittance parameters on-line based on the tracking error to compensate the unknown trajectory deviations. In the end, the developed control scheme is experimentally tested on a dual-arm setup composed of two ESTUN industrial manipulators carrying a common object. The simulations and experimental results strongly prove that the proposed approach can achieve good position/force tracking performance.