Séminaire C3M : Seppe TERRYN - VRIJE UNIVERSITEIT BRUSSEL

  Version imprimable de cet article RSS
22 septembre 14:00 » 15:00 — Amphi Boreau

Seppe Terryn , Vrije Universiteit Brussel

Self healing robots

Résumé :
Natural agents display various adaptation strategies to damages, including damage assessment, localization, healing, compensation and recalibration. Inspired by nature, we introduce these approaches in human-man agents ; in soft robots. These new generation robots consist of a soft body, usually made from elastomers. Being soft, these bioinspired robots are very safe and adaptable. However, softness comes with a price, as they are vulnerable to a variety of damages, leading to non-sustainable solutions. In our work, we have been developing self-healing soft robots, built out of multiple reversible elastomers, crosslinked by a Diels-Alder chemistry. These multi-material robots can recover from large macroscopic cuts, punctures and tears, fatigue and delamination, increasing their robustness and lifetime. These healable soft robots are manufactured using new processing techniques, including fused filament fabrication and “assembly & binding techniques”. Heat can be provided externally, or via flexible heaters, embedded in the soft robotic body. Using Diels-Alder-based conductive composites, a flexible heater was made and integrated in a robotic finger. By applying current, this heater is able to heal itself as well as the finger. In general healing requires mild heating (70-90 °C). However, careful tuning of the network composition of Diels-Alder polymers has led to soft robots that heal autonomously at ambient conditions. Conductive Diels-Alder composites are also used in self-healing sensors for deformation and force sensing in soft robots. These sensors recover their sensory performance after healing and being multi-functional they are also used for damage detection and localization, and health monitoring. However, because viscoelasticity introduces significant nonlinearities and time-variance into the sensor response, the sensor signals are challenging to model using analytical approaches. Alternatively, we show that machine learning leads to highly sensitive and robust models for healable sensors and electronic skins, able to adapt and compensate in case of limited recovery due to non-optimal healing.

Dr. ir. Seppe Terryn received his PhD in 2019 at the Vrije Universiteit Brussel, in which he combined smart materials and robotics and developed a new multidisciplinary field “self-healing soft robots”. He is currently working as Postdoctoral researcher with a personal grant of the FWO and is managing two European projects ; FET Open SHERO on self-healing soft robots, the EU Marie Curie ITN SMART on smart materials for robots and a national AMSER on additive manufacturing of self-healing robots. In these projects, he is currently co-supervising a team of 10 PhD students working further on increasing the TRL-level of self-healing soft robots towards industrial applications, while performing fundamental research on self-healing polymers/composites, (additive) manufacturing, self-healing sensors and actuators.





ÉCOLE SUPÉRIEURE DE PHYSIQUE ET DE CHIMIE INDUSTRIELLES DE LA VILLE DE PARIS
10 Rue Vauquelin, 75005 Paris