Real world robot applications have to cope with large variations in the operating conditions due to the variability and unpredictability of the environment and its interaction with the robot. Performing an adequate control using conventional control techniques, that require the model of the plant and some knowledge about the influence of the environment, could be almost impossible. An alternative to traditional control techniques is to use an automatic learning system that uses previous experience to learn an adequate control policy. Learning by experience has been formalized in the field of reinforcement learning. But the application of reinforcement learning techniques in complex environments is only feasible when some generalization can be made in order to reduce the required amount of experience. This work presents an algorithm that performs a kind of generalization called categorization. This algorithm is able to perform efficient generalization of the observed situations, and learn accurate control policies in a short time without any previous knowledge of the plant and without the need of any kind of traditional control technique. Its performance is evaluated on the trajectory tracking control with simulated DC motors and compared with PID systems specifically tuned for the same problem.
[1]
Sean R Eddy,et al.
What is dynamic programming?
,
2004,
Nature Biotechnology.
[2]
M.A. Perez-Cisneros,et al.
Reinforcement learning neurocontroller applied to a 2-DOF manipulator
,
2001,
Proceeding of the 2001 IEEE International Symposium on Intelligent Control (ISIC '01) (Cat. No.01CH37206).
[3]
Andrew G. Barto,et al.
Combining Reinforcement Learning with a Local Control Algorithm
,
2000,
ICML.
[4]
Enric Celaya,et al.
Learning in categorizable environments
,
2000
.
[5]
Carme Torras,et al.
Visual navigation outdoors: the ARGOS project
,
2002
.
[6]
Andrew G. Barto,et al.
Reinforcement learning
,
1998
.
[7]
Enric Celaya,et al.
Implementation of a hierarchical walk controller for the LAURON III hexapod robot
,
2003
.