Course Description

The course introduces the design and control principles of dynamical robots, as exampled by the MIT mini Cheetah quadruped robot, a fabulous dynamical under-actuated mechanical system. We will introduce both theories and practices, focusing on how to project theories into designs and how to dig the fundamental principles underlying practices. "Hands-on" is the most efficient way in learning, especially if you can learn while playing.

Topics include forward and inverse kinematics, robot simulation in a python-based physical engine, nonlinear control and modern control theory, deep-mimc dynamics of robotic control with motion capture, actuator design and motor control. Hopefully, we can keep you in touch with most advanced technology development, such as AI. You can design and develop robots that are more dynamic and more agile than the current state-of-the-art later on.

Course Information

Instructor

Prof. Hongtao Wang (Email: htw@zju.edu.cn)

Teaching Assistant

Ph.D. Candidate: Anda Zhou (Email: 11824014@zju.edu.cn)

Course

Advanced Experimental Solid Mechanics (ZJU Course No. 26120280)

Part II: Hands-on Deep Dynamical Robot

Lecture & Experiments, every Wed. afternoon 13:15 pm, 3 or 4 sessions

Goal

（１）Hand on dynamical robots;

（２）Understand the design, modelling, simulation, assemble and control;

（３）Keep in touch with most advanced technology development, such as AI;

Grading

Reference

---

Fabulous online courses

Practical

MIT course: How to Make Something that Makes (almost) Anything

Website: http://fab.cba.mit.edu/classes/865.18/introduction/index.html

It has all tools for building robots.

Theoretical

MIT course: Underactuated Robotics

Website: http://underactuated.csail.mit.edu/underactuated.html

Online video: https://www.youtube.com/channel/UChfUOAhz7ynELF-s_1LPpWg

Control

（１）Course: Control bootcamp by Steve Brunton from Washington University

Online video: https://www.youtube.com/watch?v=Pi7l8mMjYVE&list=PLMrJAkhIeNNR20Mz-VpzgfQs5zrYi085m

（２）Course: Reinforcement learning by D. Silver

Online videos: http://www0.cs.ucl.ac.uk/staff/d.silver/web/Teaching.html

（３）Machine learning by Andrew Ng

Online videos: https://www.coursera.org/learn/machine-learning

Online free textbooks

Robot:

（１）Russ Tedrake, Underactuated Robotics -- Algorithms for Walking, Running, Swimming, Flying, and Manipulation

http://underactuated.csail.mit.edu/underactuated.html

（２）Marc Raibert, Legged Robots That Balance, The MIT Press (not free)

https://mitpress.mit.edu/books/legged-robots-balance

（３）开源机器人学学习指南

https://github.com/qqfly/how-to-learn-robotics/tree/backup/all-in-one

Reinforcement learning:

Sutton, Richard S., and Andrew G. Barto. Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press, 1998. ISBN: 9780262193986.

http://incompleteideas.net/book/the-book-2nd.html

Control:

Karl J. Åström and Richard M. Murray, Feedback Systems: An Introduction for Scientists and Engineers

https://www.cds.caltech.edu/~murray/amwiki/index.php/Second_Edition

Neural network and deep learning:

Michael Nielsen, Neural Networks and Deep Learning

http://neuralnetworksanddeeplearning.com/chap1.html




Lectures, Assignments, Textbook

---

The lectures, assignments, textbook, and some reference materials will be all be open and available online.

The overall repository of this course is avaliable here. The following links are arranged in order of each lecture. (Please do not try to download the whole directory using the "ZIP" button on the right, which may cause heavy load to our server.)


2019.05.05

Lectures; References; Tutorial





2019.05.08

Lectures; References; Cases; Videos


(The reference video is from number 12 to 23 )




2019.05.15


Lectures; References; Impedance Control Videos





Disclaim

---

（1）We acknowledgment all authors of all references and videos used in the course. We try hard to cite these materials correctly. If anyone feels that anything is not properly acknowledged, please let me know and we will correct it ASAP.

（2）We also note that all posted references are for the only purpose of teaching. If any copyrights owner feels that it should not be used, please let me know and we will remove the posted references and stop using them in the first time. You can send email to htw@zju.edu.cn.

（3）Our course slides are free and open to anyone who is interested in Robotics! We would like to share because we have learnt quite a lot from open source materials. We hope that you can find something interesting in our slides!