Optical power auto-alignment method with eugenics sorting for enhancing the alignment speed and robustness of...

To auto-couple optical devices, a simple but effective method must have a high success rate, fast scanning speed, and high stability. For coupling accuracy, swarm intelligence algorithms set a large number of particles to find the optimal point, which can introduce accelerated geometric errors in practical engineering. In this study, we proposed a method for auto-alignment between single-mode fibers and grating couplers using the particle swarm optimization algorithm, which introduces a chaotic mapping and eugenics mechanism. With the help of chaotic mapping and eugenics mechanisms, the scanning speed and robustness increased remarkably. A series of simulations and experiments showed that this method could increase the efficiency and robustness by 90% and 50%, respectively, compared to the basic swarm intelligence algorithm.