Decoupled-Space Control and Experimental Evaluation of Spatial Electrohydraulic Robotic Manipulators Using Singular Value Decomposition Algorithms

摘要

This paper proposes a decoupled-space control (DSC) framework for spatial multi-degrees-of-freedom (DOF) parallel robots, to further improve the control performances of the currently studied control algorithms, via realizing the decoupling of the control space, using singular value decomposition (SVD) algorithms. Making use of the positive definite property of the mass/inertia matrix of spatial multi-DOF parallel robots, the decoupled matrix, an orthogonal matrix, can be derived by applying SVD algorithms. Through space transformation with the decoupled matrix, a spatial multi-DOF parallel robot model can be expressed in an uncoupled space. The independent control design can be implemented for each control channel in the uncoupled space, and the strongly dynamic coupling effects in the current control strategies can be removed. To confirm the proposed DSC framework, a simple controller, as an example, is developed in the decoupled space with the desired and real actuator position as its input and the valve command as its output. Under the proposed DSC strategy, the control performances of spatial multi-DOF parallel robots are evaluated in simulation and experiment. Results show that the DSC framework can further improve the control performances of the current control algorithms designed in the physical space of spatial multi-DOF parallel robots, by solving the problem of dynamic coupling effects.