Reinforcement learning applied to multidisciplinary systems design optimization of an aerial vehicle

The engineering design problems have been highly complex and time-consuming. Real-world engineering systems also suffer many systems and subsystems levels challenges during the entire product life cycle because of their inevitable multidisciplinary nature and complex coupling between different subsystems. Therefore, any effort in direction of alleviating difficulties in the field of Multidisciplinary Systems Design Optimization (MSDO) will receive remarkable attention. Artificial Intelligence (AI) which is a massive field encompassing many goals has recently triggered a paradigm shift in numerous industries around the world and also could be led to a revolution in the MSDO field of research. Currently, the most influential topic in AI is machine learning (ML) which is decomposed into supervised learning, unsupervised learning, semi-supervised learning, and Reinforcement Learning (RL). The focus of this article is to show the further applications of RL in the field of MSDO research area. To demonstrate the potential capability of this strategy, it is applied to solve some optimization benchmark problems and also design optimization of an aerial vehicle.