Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.
This lecture delves into the importance of modeling and predicting uncertain environments for ensuring safe and high-performance autonomy in modern autonomous systems. The instructor, Prof. Francesco Borrelli from UC Berkeley, presents research focusing on control design for autonomous systems integrating predictions and learning while ensuring safety. The lecture covers the comparison with model-free approaches and discusses key open questions in the field. Topics include model-based predictive control, AI/ML, disciplined control design, offline and online computation, feedback policies, and addressing model mismatch. The presentation also touches on challenges and solutions for implementing predictive control in neuroscience and showcases real-world applications and experiments.