Kinematic and dynamic design and optimization of a parallel rehabilitation robot

In this paper, a method for concurrent optimum design of a complex parallel manipulator is introduced. The manipulator is a three-degree-of-freedom mechanism used as a walking rehabilitation device. The proposal deals with several optimization issues; firstly, the methodology is applied to a system recently designed and, in the best of our knowledge, the control policy, and dynamic model have not been published before, secondly, we propose an objective function which considers dexterity and singular manipulators, as well as energy and position error, and thirdly, we propose an optimization algorithm which successfully approximates the optimum solution, delivering low-cost feasible designs with fewer function evaluations than a comparing Genetic Algorithm. A set of numerical simulations validate the methodology and evidence its robustness since it delivers quite similar designs in several independent executions.