Constraint-Handling Techniques for the Concurrent Design of a Five-Bar Parallel Robot

Abstract

In this paper, a concurrent structure-control design is applied to a five-bar parallel robot in order to minimize the tracking error in a high-speed task. The high power necessary to perform the task leads to a challenging problem in handling a physical constraint on the maximum torque as well as on the structure-control design parameters. Three constraint-handling techniques, the feasibility rules, ε-constrained method and stochastic ranking have been used to analyze how the design changes as the torque constraint are reduced. Studying torque reduction is important from the energy consumption point of view, as well as the use of lighter and cheaper motors. In addition, a comparison among the three constraint-handling techniques is implemented so as to observe the performance and quality of the obtained results. The final results suggest that stochastic ranking method always obtains feasible and successful solutions in all proposed torque limits.

Keywords

• Handling torque constraints
• control-structure design
• pick and place tasks
• high-speed
• differential evolution