Lecture 13: MIT 6.832 Underactuated Robotics (Spring 2022) | "Trajectory Stabilization 2"

This lecture provides a rigorous and insightful introduction to trajectory stabilization for underactuated systems. The instructor, Russ Tedrake, is a leading expert in the field, and his expertise is evident in the clarity and depth of the presentation. The lecture begins with a compelling motivation: the perching example, where open-loop trajectory optimization fails due to numerical errors. This effectively illustrates the need for feedback control. The core concept of linearizing around a trajectory is explained intuitively, with a clear derivation of the time-varying linear system. The extension of LQR to time-varying systems is presented with mathematical rigor, including the Riccati equation. The lecture also touches on important practical considerations, such as the finite horizon and the region of attraction. The use of simulations and code examples enhances understanding. The content is well-structured and builds logically on previous lectures. The sources are not explicitly cited in the video, but the material is based on established control theory and the instructor’s own research. Overall, this is an excellent lecture that provides a solid foundation for understanding trajectory stabilization. The only minor weakness is the lack of explicit references, but this is common in lecture settings. The adéquation between title and content is perfect. The lecture is highly valuable for students and practitioners in robotics and control.

The title accurately reflects the content: a lecture on trajectory stabilization, specifically focusing on LQR for time-varying systems and its application to the perching example.

9/10

Lecture from MIT OpenCourseWare by a leading expert in robotics. Content is rigorous, well-structured, and based on established control theory. The instructor demonstrates concepts with concrete examples and simulations. No commercial bias detected.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction and motivation: perching example, open-loop trajectory optimization fails.
• 05:00Demonstration of open-loop simulation missing the perch; discussion of numerical errors.
• 10:00Introduction of LQR along a trajectory; finite-horizon LQR stabilizes the perch.
• 15:00Linearization around a trajectory: time-varying linear system, error coordinates.
• 20:00Extension of LQR to time-varying systems; Riccati equation still applies.
• 25:00Discussion of finite horizon vs infinite horizon; time-varying Q and R.
• 30:00Region of attraction for time-varying LQR; limitations of linear control.
• 35:00Conclusion and preview of next topics: robust control, nonlinear synthesis.