NRI/Collaborative Research: Robust Design and Reliable Autonomy for Transforming Modular Hybrid Rigid-Soft Robots

Texas A&M Engineering Experiment Station, College Station TX

Investigators

Abstract

Soft robots are a promising technology for exploration tasks and have shown great potential for unlocking the secrets of biological creatures and creating adaptable, versatile, and human-friendly robots. On the other hand, rigid robots offer many advantages in terms of precision and ease of control. The novel hybrid robot design considered in this project will combine these two technologies to provide the advantages of both. This grant supports research to produce new knowledge related to modular design and reliable human-robot collaboration using this hybrid design paradigm. The new modular robots will be able to adapt to new environments, move across varied terrains, and operate in a wide variety of environmental conditions. This project will result in improved robotic capability for exploration of difficult terrains such as those found during search and rescue, inspection, surveillance, and reconnaissance. The new robot capabilities will also benefit future extra-terrestrial missions. The project includes summer research opportunities for undergraduate students and educational outreach activities to K-12 students to support soft robotics education. The research supported by this grant will significantly advance the state-of-the-art in the theory and practice of soft robots by creating novel and hybrid (incorporating both stiff and soft elements) physical hardware designs to offset limitations of each individual technology. The research tasks are aimed at advancing the seamless integration and operation of non-collocated human-robot teams and are aligned in two thrusts. The first thrust considers modular design for extreme environments using low-degree-of freedom modules connected by reversible, completely sealed permanent magnet connectors. Modules communicate with each other via wireless protocols to enable distributed control algorithms. Dynamic modeling and simulation are used to discover a hybrid locomotion strategy based on a learning central pattern generator. The project also creates motion planning algorithms for bioinspired, topologically reconfigurable soft robots. The second thrust focuses on reliable non-collocated human robot collaboration, including high-level teleoperation, topological transformation strategies to mitigate module failures, automatic gait re-selection based on fault diagnosis techniques, and a consensus decision-making system to enable functional modules to disable nonfunctional ones. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.