Methodology for Parameters Optimization of an Hybrid Architecture of Control
| Authors: | Adouane Lounis, LAB CNRS 6596, France
le Fort-Piat Nadine, LAB CNRS 6596, France |
|---|
| Topic: | 3.2 Cognition and Control ( AI, Fuzzy, Neuro, Evolut.Comp.) |
|---|
| Session: | Soft Computing for Control |
|---|
| Keywords: | Behavioral architecture of control, Cooperative mobile robotics, Parameters optimization, Genetic algorithms. |
|---|
Abstract
This paper presents a methodology for parameters optimizationapplied to an hybrid behavioral control architecture. Thecoordination between behaviors in this architecture is insuredusing both hierarchical and fusion action mechanisms. This globalmechanism of coordination is characterized by a multitude ofparameters which must be finely tuned to enhance the efficiency ofthe execution of cooperative tasks. The proposed parametersoptimization is obtained using genetic algorithms. Appropriategenetic operators are used to manipulate real chromosomes with anormalization relationship between its genes. The validation ofthe results is established using a large number of simulations.
