Joint Center for Robotics

presents the 6^th Robotics Workshop

 

Advaand

- no cost to register

- snacks, beverages, and lite lunch provided

- registration is first come first serve - U.S. Citizens Only!

- non-government attendees MUST register by Sept 7, 2009 at 5:00pm

 

 

 

 

AGENDA and Abstracts

Advances in Walking Robots

Rough Terrain Locomotion

 

 

Bipedal Walking and Push Recovery

Dr. Jerry Pratt,

Institute for Human and Machine Cognition

 

Abstract: Humanoid robots are a potentially useful platform for operations in complex urban environments. They have the potential to get to many of the places that humans can get to and do many of the tasks that humans can do. To date, there are many humanoid robots that can walk on flat ground and some that can climb stairs. However, most humanoid robots fall when encountering moderate unknown terrain or when slightly pushed. In this talk, we will present bipedal walking and push recovery algorithms for our lower body humanoid robot, M2V2. The walking algorithms rely on low impedance force controllable actuation, which helps improve disturbance robustness. The push recovery algorithms utilize the concept of Capture Points. A Capture Point is a point on the ground in which a legged robot can step to and stop within one step. The Capture Region is the set of all Capture Points. In order to recover from a push, the robot estimates the Capture Region and steps into it. We will present simple models for estimating Capture Points, present a robustness metric based on Capture Points, and show how Capture Points can be used for motion planning over stepping stones.

 

back to top

 

 

 

 

 

 

 

 

BigDog, the rough-terrain robot - What's done, what's next?

Dr. Marc Raibert & Dr. David Wooden,

Boston Dynamics

 

Abstract: BigDog is a quadruped robot designed to operate in rough terrain. It has hiked in rocky, muddy and snowy terrain, traveled 12 miles without stopping, followed a human leader visually, and done limited autonomous cross country travel. This talk will give an overview of BigDog technology, describe progress so far, and outline plans for BigDog's next steps.

back to top

 

 

 

 

 

 

 

 

Motion Planning and Feedback Control for Rough Terrain Locomotion

Dr. Russ Tedrake, 

Massachusetts Inst. of Technology

 

Abstract coming soon.

 

back to top

 

 

 

 

 

 

 

 

Passive Dynamic Walking

Dr. Art Kuo

University of Michigan

 

Abstract coming soon.

 

back to top