abstract
- © 2020, Springer Nature Switzerland AG.We introduce an optimization algorithm which combines an Estimation of Distribution Algorithm (EDA) and the Nelder-Mead method for global and local optimization, respectively. The proposal not only interleaves global and local search steps but takes advantage of the information collected by the global search to use it into the local search and backwards, providing of an efficient symbiosis. The algorithm is applied to the concurrent optimization of a rehabilitation robot design, that is to say, to the dimensional synthesis as well as the determination of control gains. Finally, we present an statistical analysis and evidence about the performance of this symbiotic algorithm.