| description abstract | Hybrid mechanisms combine the advantages of serial mechanisms and parallel mechanisms and have better stiffness and a larger workspace. In this paper, a novel workspace strongly coupled machining robot mechanism with a hybrid module is proposed. It is composed of a four-degrees-of-freedom (4DOF) R(3-RUHR)/UURP hybrid module and a 2-DOF (RRRP)R single-closed-loop serial rotation module, which has the advantages of small footprint, large workspace, high stiffness, and is ideal for machining robots applied to modular integrated manufacturing lines. However, due to the multi-closed loop between azimuth pitch rotations and hybrid module, the workspace is coupled, so that the mechanism has problems in building the forward and inverse position, kinematics, and dynamics analytical models, numerical solution calculation speed, and multi-solution. Based on the analysis of coupled characteristics of this mechanism, the concomitant motion neural network (NN) models of the 3-RUHR parallel mechanism and inverse position NN models of the (R(3-RUHR)/UURP)(RRRP)R hybrid robot are established using the Levenberg–Marquardt training algorithm with Bayesian regularization. Furthermore, the kinematics and dynamics models of the proposed hybrid robot are established using the equivalent screw method. And the correctness of position NN models, kinematics complete model, and dynamics complete model are verified through numerical examples, respectively. Finally, the prototype of this hybrid robot is further developed, and the coupled motion and multi-axis machining capability of the (R(3-RUHR)/UURP)(RRRP)R hybrid machining robot are verified by experiment. The hybrid mechanism and modeling method proposed in this paper can provide support for the application of machining robots with a small footprint, strongly coupled large workspace, and high stiffness. | |
