6.832 - Underactuated Robotics - MIT Lecture Notes

Instructor(s) - Prof. Russell Tedrake

MIT Course Number - 6.832

Level - Graduate

Robots today move far too conservatively, using control systems that attempt to maintain full control authority at all times. Humans and animals move much more aggressively by routinely executing motions which involve a loss of instantaneous control authority. Controlling nonlinear systems without complete control authority requires methods that can reason about and exploit the natural dynamics of our machines. This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

Course Notes table.

CHAPTERS	TOPICS
Front	Title page, table of contents, and preface (PDF)
1	Fully actuated vs. underactuated systems (PDF)
I. Nonlinear dynamics and control
2	The simple pendulum (PDF)
3	The acrobot and cart-pole (PDF)
4	Manipulation
5	Walking (PDF)
6	Running
7	Flight
8	Model systems with stochasticity
II. Optimal control and motion planning
9	Dynamic programming (PDF)
10	Analytical optimal control with the Hamilton-Jacobi-Bellman sufficiency theorem (PDF)
11	Analytical optimal control with Pontryagin's minimum principle
12	Trajectory optimization (PDF)
13	Feasible motion planning
14	Global policies from local policies
15	Stochastic optimal control
16	Model-free value methods
17	Model-free policy search (PDF)
18	Actor-critic methods
IV. Applications and extensions
19	Learning case studies and course wrap-up
Appendix	A. Robotics preliminaries (PDF) B. Machine learning preliminaries
Back	References (PDF)

Prof. Russell Tedrake, 6.832, Underactuated Robotics. 

(Massachusetts Institute of Technology: MIT OpenCouseWare), http://ocw.mit.edu (Accessed August 27, 2013). 

License: Creative Commons BY-NC-SA

Our website abides by the Creative Commons BY-NC-SA as set by MIT.