Increasing levels of automation are beneficial for safety and efficiency of systems such as cooperative and automated vehicles, robotics, or powertrains. To enable this increased automation in complex environments, this research will focus on the intersection of classical control theory and artificial intelligence (in particular, machine learning and optimization). With increasingly more sensors and data recorded in all existing devices, new opportunities are created to improve how we design and control newly developed systems. Recent studies show that creating a bridge and a hybridization between AI methods and control theory leads to improving system’s performance (be that servo performance, tracking error, energy efficiency, reaction times, etc..) and overall autonomy. To enable these developments, our research combines offline learning and online prediction of behaviors (e.g., driving, cycling, walking), with context and world modelling, to improve controller, planning and decision-making synthesis and functional safety in the area of cooperative and automated mobility (e.g., passenger vehicles, trucks) and mobile robots.