Recently, many researchers are focusing their attention on the Evolutionary Computation (EC) methods. EC can provide good solutions for complicated optimization problems although it requires a lot of iterations of calculation. One of the solutions to this problem is to perform EC in parallel. When EC is implemented to parallel computers, it is important to design a parallel model suitable for the architecture of parallel computers. In this report, the parallelization of the genetic algorithms and the simulated annealing are mentioned here.
This paper describes a method and a process for developing a soft artificial hand similar to a real human hand, which is actuated by the soft actuator, called "Mckibben pneumatic artificial muscle". The pneumatic artificial muscle makes an artificial hand not only move softly and powerfully by pneumatic power but also work as a sensor. Applying pneumatic artificial muscles, the hand becomes very soft in material as well as motion and reacts sensitively to the load for its self-sensing function in the actuator.