Robot Learning and Control

This lecture introduces the concepts of learning and adaptive control for robots, focusing on methods by which robots can learn control laws enabling real-time reactivity using dynamical systems. The course covers topics such as traditional planning approaches in robotics, path planning in 2D and global path planning. It also delves into challenges faced by real-time planners, such as highly non-linear dynamics and the need for real-time planning in dynamic environments. The lecture explores solutions like optimal control and path planning using dynamical systems, emphasizing the importance of stability guarantees and obstacle avoidance. Practical sessions involve practice on real robots to apply the theoretical concepts learned in the course.