Authors:
Laetitia Chapel
1
and
Guillaume Deffuant
2
Affiliations:
1
Université de Bretagne Sud, France
;
2
Cemagref, France
Keyword(s):
Viability theory, Capture basin, Optimal control, Support Vector Machines.
Related
Ontology
Subjects/Areas/Topics:
Adaptive Signal Processing and Control
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Machine Learning in Control Applications
;
Optimization Algorithms
;
Signal Processing, Sensors, Systems Modeling and Control
Abstract:
We propose a new approach to solve target hitting problems, that iteratively approximates capture basins at successive times, using a machine learning algorithm trained on points of a grid with boolean labels. We consider two variants of the approximation (from inside and from outside), and we state the conditions on the machine learning procedure that guarantee the convergence of the approximations towards the actual capture basin when the resolution of the grid decreases to 0. Moreover, we define a control procedure which uses the set of capture basin approximations to drive a point into the target. When using the inner approximation, the procedure guarantees to hit the target, and when the resolution of the grid tends to 0, the controller tends to the optimal one (minimizing time to hit the target). We use Support Vector Machines as a particular learning method, because they provide parsimonious approximations, from which one can derive fast and efficient controllers. We illustrat
e the method on two simple examples, Zermelo and car on the hill problems.
(More)