Dynamic Performance Enhancement and Singularities Suppression by the use of Kinematic Redundancy in Parallel Kinematic Manipulators
João Cavalcanti Santos1; Maíra Martins da Silva1
1 Universidade de São Paulo
doi:10.20906/CPS/CON-2016-0287
Resumo
Parallel kinematic machines present several advantages over serial architectures as higher rigidity, improved speed/torque ratio and better load capacity. However, they may suffer from singularity issues and limited workspace. The reconfiguration capability provided by kinematic redundancy can be seen as a promising alternative to reduce these drawbacks. On the other hand, its implementation requires a considerable increase of the mass and number of moving parts. A numerical analysis is proposed to assess the effectiveness of adding one or more level of kinematic redundancies to reduce the occurrence of singularity, enlarge the workspace and improve the speed/torque ratio. The non-redundant planar 3RRR manipulator is used as the basic configuration. The understudy redundant manipulators are characterized by the addition of one, two or three prismatic active joints, yielding the (P)RRR+2RRR, 2(P)RRR+RRR and 3(P)RRR planar manipulators. Based on Lagrangian formulation of mechanics, a complete rigid-body model is presented (including Inverse and Forward Dynamics) and a proper numerical method capable of integrating the resulting DAE problem is exposed. Optimization strategies are proposed to enable the reconfiguration capability of the redundant manipulators. Gathering all analytical methods, simulations considering control systems and disturbances can be established. The consequent study compare influences of the system characteristics, mainly the degree of redundancy. According to numerical results, one can conclude that kinematic redundancy can be an alternative to remove singularities of the workspace and improve the system dynamic performance. Nevertheless, this improvement can only be achieved by the selection of a proper optimization strategy.
Palavras-chave: Parallel Manipulators; Kinematic Redundancy; Singularities; Dynamic Performance