Control performance improvement of a parallel robot via the design for control approach

Parallel structure robots have been receiving growing attentions from both academia and industries in recent years. This is due to their advantages over serial structure robots, such as high stiffness, high motion accuracy and high load-structure ratio. Control of a parallel robot, however, renders a difficult problem to control engineers. To obtain the same degrees of freedom (DOF), a parallel structure is more complex than a serial one, and so is its dynamic model in general. To effectively control a complex mechanical structure for precise and fast performance, an advanced controller embedded with the system's dynamic model is usually desired. In cases of controlling parallel robots, however, the intensive computation due to the complexity of the dynamic model can result in difficulties in the physical implementations of the controllers for high-speed performance. To avoid heavy computation, simplified dynamic models can be obtained by applying simplification techniques, nevertheless, performance accuracy will suffer due to modeling errors. This paper suggests applying a general mechatronics design approach, i.e., the design for control (DFC) approach, to handle this problem. The underlying idea of the DFC approach is that, no matter how complex a system is, as long as its mechanical structure can be judiciously designed such that it can result in a simple dynamic model, a simple control algorithm may be good enough for a satisfactory control performance. As such, complicated controller design can be avoided, on-line computation load can be reduced and better control performance can be achieved. Through out the discussion in the paper, the design and control of a two DOF parallel robot is studied as an illustration example. Resulted control performances of several different mechanical designs derived from the same robot structure topology demonstrate the effectiveness of the DFC approach. (C) 2004 Elsevier Ltd. All rights reserved.