Data-driven discovery is currently revolutionizing how we model, predict, and control complex nonlinear dynamic systems. This course aims to discuss many existing data-driven tools and their application in the modeling and control in mechanical engineering applications. Representative data-driven methods, including supervised/unsupervised learning, reinforcement learning, balanced truncation, proper orthogonal decomposition, principal component analysis, etc., will be introduced with particular case studies. The course aims to help the students to develop a data-driven perspective to analyze and control nonlinear and complex dynamic systems, in addition to conventional physics-based models and linear control theories. The students will have the opportunity to have individual course projects to practice the data-driven modeling and control methods introduced in the class.