Scientific Program

The importance of the notion of SoS’s is now essential for tackling new application changes related to major societal issues like transportation (smart communicating vehicles, optimizing multimodal transportation systems), the environment (monitoring systems) or health care (patient monitoring and diagnosis systems, designing microsystems based on nano-micro technologies applied to biological systems). It must deal with concrete real-world problems based on integrated hardware and/or software platforms that make it possible to confirm and demonstrate the relevance of the scientific findings.

The MS2T project in particular will be based on the following experimental systems :
A set of terrestrial autonomous and communicating vehicles and a fleet of aerial mini-drones integrated into the ROBOTEX excellence project.

Additionally, the MS2T LABEX will rely on multidisciplinary projects already launched as part of the SHIC research Federation, several of which are directly related to the SoS issue. In particular, the ”SUPGEST” and ”ICARE-U” projects relate to supervision of movements for rehabilitation, aided diagnostic and remote patient monitoring using biomechanical models supplied with data connected with a set of wireless sensors embedded in a portion of the musculoskeletal system.

SCIENTIFIC FIELDS

1. Interaction and cooperation between systems - Manager : M. Bouabdallah

Managing TSoS’s presumes that interaction and cooperation mechanisms have been set up between autonomous systems connected in a network, so that the overall metasystem handles some tasks with maximum effectiveness. The scientific themes mainly concern :

1. 1. The management of information flows, particularly taking into account the changing nature of the network’s configuration and restrictions specific to each subsystem (power capacity, for example)
2. 2. The design of distributed algorithms for carrying out the various functions robustly and in a collaborative manner (perception, learning, control)
3. 3. The autonomy of interacting systems

2. Managing uncertainties - Manager : T. Denoeux

Different types of uncertainties are present at all stages of modelling a complex system : random uncertainties related to the variability of the system’s inputs and, starting at a certain level of complexity, the variability of the state of the system itself, which can only be described stochastically ; and epistemic uncertainties due to insufficient knowledge of the system (models) and of its environment (quantitative and qualitative limits of the sensors). The main themes that will be studied are :

1. 1. Modelling uncertainties and their propagation within information processing, from perception to decision-making
2. 2. Uncertainty robustness and integrity. One major issue relates to the design of algorithms that are both robust with respect to uncertainties and have high integrity, meaning that they provide results with an associated guaranteed risk level.
3. 3. Accounting for uncertainties in numerical modelling.

3. Optimized design of technological SoSs - Manager : P. Villon

The various industrial simulation tools developed over the past few years have had a limited impact on the design of complex systems made up of several subsystems. While tools for modelling each subsystem are gradually becoming refined, the general optimization of the systems has run up against methodological, organizational, informational, and numerical difficulties. This field aims to build methods and tools for the following themes :

1. 1. Multi-level and multi-physical optimization of a set of complex systems
2. 2. Scheduling and synchronizing mobile subsystems
3. 3. Operational safety of technological SoSs.