Download:
|
by Olac Fuentes, Al C. Nelson
Machine Learning
ftp://ftp.cs.rochester.edu/pub/u/nelson/1996_us_japan.ps.gz
Add To MetaCart
Abstract:
We present a method for autonomous learning of dextrous manipulation skills with robot hands. We use heuristics derived from observations made on human hands to reduce the degrees of freedom of the task and make learning tractable. Our approach consists of learning and storing a few manipulation primitives for a few prototypical objects and then using an associative memory to obtain the required parameters for new objects and/or manipulations. The parameter space of the robot is searched using a modified version of the evolution strategy. Our system does not rely on simulation; all the experimentation is performed by a physical robot, the 16-degree-of-freedom Utah/MIT hand. Experimental results show that accurate dextrous manipulation skills can be learned in a short period of time.
Citations
|
2489
|
Induction of Decision Trees
– Quinlan
- 1986
|
|
1397
|
Dynamic Programming
– Bellman
- 1957
|
|
465
|
Evolutionsstrategie: Optimierung Technischer Systeme nach Prinzipien der Biolgischen Evolution. Stuttgart-Bad Cannstatt: Verlag Frommann-Holzboog
– Rechenberg
- 1973
|
|
381
|
Numerical Optimization of Computer Models
– Schwefel
- 1981
|
|
182
|
Learning to coordinate behaviors
– Maes, Brooks
- 1990
|
|
132
|
Reward functions for accelerated learning
– Mataric
- 1994
|
|
34
|
Coordinated control programs for movements of the hand
– Arbib, A, et al.
- 1985
|
|
26
|
Learning reliable manipulation strategies withoutinitial physical models
– Christiansen, Mason, et al.
- 1991
|
|
25
|
The Nature of Human Prehension: Three Dextrous Hands in One
– Iberall
- 1987
|
|
23
|
Design of the Utah/MIT Dextrous Hand
– Jacobsen, Iversen, et al.
- 1986
|
|
17
|
Aquiring visual-motor models for precision manipulation with robot hands
– Jagersand, Nelson
- 1996
|
|
16
|
Evolutionsstrategien zur numerischen Losung von Adaptationsaufgaben
– Born
- 1978
|
|
14
|
Learning to grasp using visual information
– Kamon, Flash, et al.
- 1996
|
|
14
|
Learning social behavior
– Mataric
- 1997
|
|
12
|
Global convergence of genetic algorithms: an infinite Markov chain analysis
– Eiben, Aarts, et al.
- 1991
|
|
12
|
Primitive based control of the Utah/MIT dextrous hand
– Speeter
- 1991
|
|
8
|
Complaint manipulation with a dexterous robot hand
– Michelman, Allen
- 1993
|
|
8
|
Simulated Annealing and Bolzmann Machines
– Aarts, Korst
- 1989
|
|
7
|
Morphing hands and virtual tools (or what good is an extra degree of freedom
– Fuentes, Nelson
- 1994
|
|
6
|
Automatic Discovery of Robotic Grasping Configurations
– Dunn, Segen
- 1988
|
|
6
|
Experiments on dextrous manipulation without prior object models
– Fuentes, Nelson
- 1996
|
|
6
|
Dexterous Robot Hands: Kinematics and Control
– Narasimhan
- 1988
|
|
6
|
The Virtual Tool Approach to Dextrous Telemanipulation
– Fuentes, Nelson
- 1996
|
|
6
|
Fuentes Virtual Tools: A Framework for Simplifying SensoryMotor
– Nelson, Jagersand, et al.
- 1995
|
|
5
|
An investigation of a genetic algorithm in continuous parameter space
– Grossman, Davidor
- 1992
|
|
3
|
Olac Fuentes, "Virtual Tools: A Framework for Simplifying Sensory-Motor Control in Complex Robotic Systems
– Nelson, Jagersand
- 1995
|
|
3
|
Active Exploration Based ID-3 Learning for Robot Grasping
– Salganicoff, Kunin, et al.
- 1994
|
|
1
|
An investigation of a genetic algorithm in continuousparameter space
– Davidor
- 1992
|
