Home
People
Publications
Talks/Lectures
Video Highlights
Links
Directions

The HAWK Seminar Series

Neural Network Control Project

Positions Available

The goal of our research is to build machines which exploit their natural dynamics to achieve extraordinary agility and efficiency. We believe that this challenge involves a tight coupling between mechanical design and underactuated nonlinear control, and that tools from machine learning and optimal control can be used to produce this coupling when classical control techniques fail. Our projects include minimally-actuated dynamic walking on moderate terrain, quadrupedal locomotion on extreme terrain, fixed-wing acrobatics, flapping-winged flight, and feedback control for fluid dynamics.

The Robot Locomotion Group is a part of the CSAIL Center for Robotics.

Locomotion Group Paper and Multimedia News  

Motor Learning at Intermediate Reynolds Number: Experiments with Policy Gradient on the Flapping Flight of a Rigid Wing
by John W. Roberts and Lionel Moret and Jun Zhang and Russ Tedrake

This work describes the development of a model-free reinforcement learning-based control methodology for the heaving plate, a laboratory experimental fluid system that serves as a model of flapping flight. Through an optimized policy gradient algorithm, we were able to demonstrate rapid convergence (requiring less than 10 minutes of experiments) to a stroke form which maximized the propulsive efficiency of this very complicated fluid-dynamical system. This success was due in part to an improved sampling distribution and carefully selected policy parameterization, both motivated by a formal analysis of the signal-to-noise ratio of policy gradient algorithms. The resulting optimal policy provides insight into the behavior of the fluid system, and the effectiveness of the learning strategy suggests a number of exciting opportunities for machine learning control of fluid dynamics.

Final version now available.

{LQR}-{T}rees: Feedback motion planning on sparse randomized trees

by Russ Tedrake

Recent advances in the direct computation of Lyapunov functions using convex optimization make it possible to efficiently evaluate regions of stability for smooth nonlinear systems. Here we present a feedback motion planning algorithm which uses these results to efficiently combine locally-valid linear quadratic regulator (LQR) controllers into a nonlinear feedback policy which probabilistically covers the reachable area of a (bounded) state space with a region of stability, certifying that all initial conditions that are capable of reaching the goal will stabilize to the goal. We demonstrate the performance of this systematic nonlinear feedback control design algorithm on the model underactuated systems and discuss the potential for control of more complicated control problems like bipedal walking.

RSS 2009 final version now available, with improved LTV verification.

Stable Dynamic Walking over Rough Terrain: Theory and Experiment

by Ian R. Manchester and Uwe Mettin and Fumiya Iida and Russ Tedrake

We propose a constructive control design for stabilization of non-periodic trajectories of underactuated mechanical systems. An important example of such a system is an underactuated 'dynamic walking' biped robot walking over rough terrain. The proposed technique is to compute a transverse linearization about the desired motion: a linear impulsive system which locally represents dynamics about a target trajectory. This system is then exponentially stabilized using a modified receding-horizon control design. The proposed method is experimentally verified using a compass-gait walker: a two-degree-of-freedom biped with hip actuation but pointed stilt-like feet. The technique is, however, very general and can be applied to higher degree-of-freedom robots over arbitrary terrain and other impulsive mechanical systems.

ISRR 2009 final version now available.

Locomotion Group News  

September 27, 2009. Award. Russ Tedrake was awarded the 2009 DARPA Young Faculty Award.

August 2, 2009. Positions Available The Robot Locomotion Group currently has a number of positions available.

July 14, 2009. Katie Byl has accepted an Assistant Professor position at UC Santa Barbara. Congratulations Katie!

July 1, 2009. Best Paper Award. The LQR-Trees paper won the Best Paper Award at Robotics: Science and Systems, 2009. A MATLAB software distribution for the LQR-Trees algorithm is coming soon.

June 11, 2009. Dynamic Walking 2010. The Dynamic Walking meeting will be held a MIT in 2010, from July 8-11, organized by Russ Tedrake and Hugh Herr.

May 8, 2009. Congratulations to Elena Glassman for winning the EECS Masterworks Oral Thesis Presentation Award.

April 10, 2009. Congratulations to Woody Hoburg and Elena Glassman for winning the NSF Graduate Fellowship.

March 1, 2009. Fumiya Iida has accepted a SNF Assistant Professorship at ETH Zurich. Congratulations Fumiya!

February 1, 2009. Thesis. Congratulations to John Roberts for passing the qualifiers and submitting his Masters thesis!.

October 1, 2008. We've Moved. The Robot Locomotion Group has moved to the third floor.

September 22, 2008. Robotics Challenges for Machine Learning Workshop. The full-day workshop at IROS 2008 was a great success. PDFs of many of the talks and all of the poster abstracts are available on the meeting website. We plan to organize another Robot Learning workshop in 2009.

August 28, 2008. Thesis. Katie Byl successfully defended her thesis. Congratulations Katie!

August 22, 2008. EFRI. Russ Tedrake, Sebastian Seung, Alex Megretski, and Hongkun Park have been awarded an NSF EFRI for investigating the dynamics of neural networks on a planar patch-clamp array. You can find NSF's press release here.

August 5, 2008. The Phoenix (a 2m wingspan ornithopter) has taken it's first autonomous flight. Photos by Jason Dorfman.

  

July 26, 2008. Katie Byl was awarded the IFRR Student Fellowship for her paper/presentation at ISER 2008

July 3, 2008. The MIT Learning Locomotion (LittleDog) Team was awarded phase III funding. Photo by Jason Dorfman.



June 12, 2008. Russ Tedrake was awarded an NSF Career Award

May 28, 2008. Elena Glassman was awarded an NDSEG Graduate Fellowship

May 25, 2008. Technical Committee on Robot Learning. The new IEEE Technical Committee on Robot Learning is now online.

May 18, 2008. Russ Tedrake was awarded the Jerome Saltzer award for undergraduate teaching