|
|
Foundations of Robotics
Seminar, November 15, 2005
Time
and Place | Seminar Abstract | Speaker
Biography | Presentation Slides | Speaker
Appointments
Design and Control
of an Autonomous Spider-Like Robot for Motion in 2D Tunnel Environments
Amir Shapiro
1507 Newell-Simon Hall
Refreshments 4:15 pm
Talk 4:30 pm
In conventional motion planning a wheeled mobile
robot navigates toward a goal configuration while avoiding collision with
obstacles. However, many motion-planning problems are more suited for legged
robots that interact with the environment in order to achieve stable
locomotion. For example, surveillance of collapsed structures for survivors,
inspection and testing of complex pipe systems, and maintenance of hazardous
structures such as nuclear reactors, all require motion in congested,
unstructured, and complex environments. In this work a second generation of
three-legged planar spider-like robot has been developed. The spider-robot is
capable of moving in wide range of planar tunnel environments. The robot moves
in a quasi-static manner, by stably bracing itself against the tunnel walls
while moving its free parts to the next position. We present an algorithm,
called PCG, for planning the foothold positions of spider-like robot in planar
tunnels bounded by piecewise linear walls. Using efficient convex programming
and graph search techniques, the algorithm generates a 3-2-3 gait pattern that
moves the robot from start to target using a minimum number of steps. A control method for this class of robots
will be discussed. The control method ensures that when a spider-like mechanism
bracing against the environments at equilibrium posture the naturally occurring
compliance at the contacts stabilizes the mechanisms as a single rigid body.
Finally, experimental results as well as a
short video of the spider robot motion in tunnel environment will be presented.
|
Speaker Appointments |
For appointments, please contact Amir
Shapiro (ashapiro@bgu.ac.il)
The Robotics Institute is part of the School of Computer Science, Carnegie Mellon University.