Natural creatures, from fish and cephalopods to snakes and birds, combine neural control, sensory feedback and compliant mechanics to effectively operate across dynamic, uncertain environments. To fully capture how these natural creatures interact with their environment, as well as design control methods for soft, complaint structures, it is necessary to incorporate the role that soft-matter physics have on these structures.
Elastica can serve as a versatile platform within which novel control methodologies for such soft structures may be developed. Currently most control algorithms are based on rigid body dynamics, however, soft body structures introduce new dynamics that rigid body solvers are not able to account for. By developing control algorithms with Elastica, these soft body dynamics can be incorporated and exploited for more precise and robust control.
Elastica is being used to enable control in multiple different projects.
This ONR MURI project seeks to develop a simulation environment where new control algorithms based on the neurodynamics and biology of octopuses can be used to enable control of the next generation of soft robotics. Elastica is being used to enable accurate simulation of the octopuses arms. Learn more at the project website.
NCSA-NVIDIA AI Hackathons are co-sponsored by the National Center for Supercomputing Applications (NCSA) and NVIDIA. The goal of the hackathons are to let talented University of Illinois students, postdocs and staff showcase their skills while working on challenging problems involving deep learning on a state-of-the-art compute platform designed for AI. Elastica has been used to pose challenging problems for the Hackathon related to developing AI control algorithms for soft-body structures.
Elastica + Control API
Based on the work being done in the CyberOctopus project and the NCSA-NVIDIA Hackathons, we are working to develop an Elastica + Control API that is compatible with OpenAI Gym to enable an easy interface for researchers working on control who wish to use Elastica.
Elastica + Control Publications
Naughton, Noel, Jiarui Sun, Arman Tekinalp, Girish Chowdhary, and Mattia Gazzola, Elastica: A compliant mechanics environment for soft robotic control, arXiv:2009.0842.
Chang, Halder, Shih, Tekinalp, Parthasarathy, Gribkova, Chowdhary, Gillette, Gazzola, Mehta, Energy shaping control of a CyberOctopus soft arm, arXiv:2004.05747.