Thadeu TUCCI : "Distributed Algorithms for Programmable Matter..."

PhD work : "Distributed Algorithms for Programmable Matter: Target Shape Description and Self-Assembly Planning"

Abstract :

Programmable matter can be seen as a huge modular robot in which each module can communicate to its connected neighbors and work all together to achieve a common goal, more likely changing the shape of the whole robot and adapting it with new functionalities. In order to achieve coordination between a group with many thousand robots, local rules and distributed algorithms would take advantage in this environment. In the same way, small modules means there is also small resources and algorithms should be designed to reflect these needs. This thesis provide algorithms and solutions to solve some parts of the self-reconfiguration problem with each module embedding the same algorithm and coordinating with the others by means of neighbor-to-neighbor communication. One of them is a study and proposal of a representation for the goal structure that reduces the footprint memory. Also, the self-assembly like self-reconfiguration is composed of two steps: (1)

identifying the free positions that are available for docking and (2) moving and docking modules to these positions. In this thesis, distributed solutions for the first step are presented which can decide positions that can be filled and can create any 3D shape, including shapes with internal holes and concavities.

Composition of the jury :

Mme. Marie-Pierre GLEIZES, Full Professor at University of Toulouse, Reporter
Olivier SIMONIN, Full Professor at Université of Lyon, Reporter
Gauthier PICARD, Full Professor at  Ecole Nationale Supérieure des Mines de Saint-Etienne, Examiner
Julien BOURGEOIS, Full Professor at Université Bourgogne - Franche-Comté, PhD Supervisor
Benoit PIRANDA, Associate Porfessor at Université Bourgogne - Franche-Comté, PhD Co-Supervisor