Reinforcement Learning for Control Design

Learn how to design model-free controllers using Reinforcement Learning techniques in this 19-minute conference talk that bridges the gap between traditional control systems theory and modern machine learning approaches. Explore the application of RL methods to create Linear Quadratic Regulator (LQR) controllers without requiring explicit system models, moving beyond the conventional model-based approaches that have dominated the mature field of control systems. Discover how this innovative approach can be applied across diverse control applications while maintaining the effectiveness of traditional control design principles.