A Nervousness Regulator Framework for Dynamic Hybrid Control Architectures

José-Fernando Jimenez Gordillo (1) , Abdelghani Bekrar (2) , Damien Trentesaux (2) , Paulo Leitão (3, 4)

1 PUJ - Pontificia Universidad Javeriana
2 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201
3 Polytechnic Institute of Bragança
4 LIACC - Departamento de matematica aplicada [Porto]

Dynamic hybrid control architectures are a powerful paradigm that addresses the challenges of achieving both performance optimality and operations reactivity in discrete systems. This approach presents a dynamic mechanism that changes the control solution subject to continuous environment changes. However, these changes might cause nervousness behaviour and the system might fail to reach a stabilized-state. This paper proposes a framework of a nervousness regulator that handles the nervousness behaviour based on the defined nervousness-state. An example of this regulator mechanism is applied to an emulation of a flexible manufacturing system located at the University of Valenciennes. The results show the need for a nervousness mechanism in dynamic hybrid control architectures and explore the idea of setting the regulator mechanism according to the nervousness behaviour state.

Mots clés: Nervousness; Dynamics; Hybrid control architecture; Switching; Multi-agent system

Domaines: Automatique / Robotique

José-Fernando Jimenez Gordillo, Abdelghani Bekrar, Damien Trentesaux, Paulo Leitão. A Nervousness Regulator Framework for Dynamic Hybrid Control Architectures. Service Orientation in Holonic and Multi-Agent Manufacturing, 640, Springer International Publishing, pp.199-209, 2016, Studies in Computational Intelligence, ⟨10.1007/978-3-319-30337-6_19⟩. ⟨hal-03411575⟩