The authors consider the problem of a robot manipulator operating in a noisy workspace. The manipulator is required to move from an initial position P/sub i/ to a final position P/sub f/. P/sub i/ is assumed to be completely defined. However, P/sub f/ is obtained by a sensing operation and is assumed to be fixed but unknown. The authors approach to this problem involves the use of three learning algorithms, the discretized linear reward-penalty (DL/sub R-P/) automaton, the linear reward-penalty (L/sub R-P/) automaton and a nonlinear reinforcement scheme. An automaton is placed at each joint of the robot and by acting as a decision maker, plans the trajectory based on noisy measurements of P/sub f/. >
[1]
Mitsuo Kawato,et al.
Feedback-error-learning neural network for trajectory control of a robotic manipulator
,
1988,
Neural Networks.
[2]
John J. Craig,et al.
Introduction to Robotics Mechanics and Control
,
1986
.
[3]
S. Arimoto,et al.
Learning control theory for dynamical systems
,
1985,
1985 24th IEEE Conference on Decision and Control.
[4]
Suguru Arimoto,et al.
Bettering operation of Robots by learning
,
1984,
J. Field Robotics.
[5]
Charles P. Neuman,et al.
Robust nonlinear feedback control for robotic manipulators
,
1985
.