The DiCOT team performs theoretical, methodological and applied researchs to design fault tolerant controlled systems. This research concerns the design of Fault Detection and Isolation (FDI) systems, Fault tolerant Control (FTC) systems, and Fault Tolerant Fusion (FTF).
The strength of the DiCOT team is to gather researchers on automatic control, dependability analysis, data analysis and data fusion from different scientific communities. This wealth enables us to tackle the fault tolerance problems according to various facets. It enables us to develop original and efficient methods which allowus to take advantage of different specialities and to tackle the fault tolerance problem from a system point of view. We mainly consider linear and nonlinear continuous-time systems and hybrid dynamical systems (switching systems). The goals of our research team are to study and take into account the interactions between the FDI system and the different algorithms used in controlled systems. Active Fault Tolerant Control (AFTC) uses a Fault Detection and Isolation (FDI) module to adapt the control law in order to react to faults through reconfiguration or accommodation (adaptive control). The goals are to maintain the overall system stability and acceptable performance in spite of the occurrence of faults. When integrating the different units to build a reliable AFTC law it is generally considered that each individual subsystem operates correctly: its output is instantaneously available to provide decisions/actions to other subsystems. The interactions between the reconfigurable controller and the FDI unit are not fully considered. From another point of view, control schemes need generally observation, estimation and data fusion modules. These modules must also be fault tolerant in order to guarantee global system performances. The FDI algorithms have also been used to reconfigure them when faults occur. The main application areas concern the multi-robot systems, transportation systems and energy systems.
Two themes will be considered in the DiCOT team :
Maan El Badaoui El Najjar
Surveillance des réseaux embarqués des véhicules autonomes : une approche conjointe signal et système
Caractérisation fine de l'environnement de transport par traitement d'images : contribution à la localisation multi-sources d'un véhicule autonome
Méthodologie orientée sûreté de fonctionnement pour la cyber-sécurité des systèmes de contrôle-commande
Approches informationnelles pour une navigation autonome colaborative de robots d'exploration de zones à risques
Tolérance aux fautes pour la navigation autonome d'un véhicule
Localisation tolérante aux fautes par couplage serré de mesure GNSS, capteurs proprioceptifs et cartographie 3D