RI: Small: Collaborative Research: A Modular Approach to Robot Systems Incorporating Compliant and Soft Elements

Biological systems can demonstrate impressive energy efficient manipulation, from picking up fragile objects to lifting heavy weights. To do so, they rely on building blocks of varying stiffness, such as bones and muscles. Inspired by these biological systems, this project is investigating the combination of "stiff" and "soft" robotic components, creating novel "hybrids" that exploit the advantages, and conceal the inherent weaknesses, of the individual technologies. Such "hybrids" aim to bring safe and practical soft robotics to previously inaccessible domains and applications, including urban search/rescue, disaster relief, and assistance for the elderly and people with disabilities. The project will also promote robotics among underrepresented groups at both DePaul and Clemson. The project is collaboratively designing, modeling, creating, and demonstrating a series of innovative soft, modular robots. The modules are exploiting the hard/soft component interaction and constrained actuator arrangement to control stiffness independently from bending. Exploiting modularity, these robots can reassemble, or "evolve", into different manipulators and locomotors. The project is developing a unified design and optimization framework to optimize, per user specifications, the composition of hard/soft modules and optimal sensor placement. Resulting modules are fabricated and assembled into a variety of robotic configurations to demonstrate the importance of stiffness modulation in dealing with unforeseen circumstances. Using these modules, the project is also creating a sophisticated physical platform for providing new theoretical insights and innovative operational modes for soft modular robots, as well as demonstrating the fundamentals of morphological computation. Morphological computation utilizes physical body properties (e.g., stiffness and bending shape) as a computational resource that can share the burden of control to optimize stability and locomotion efficiency in real-time. This thrust is exploiting the intrinsic resonant motions of modular systems, via stiffness and shape modulation, to minimize energy loss and improve stability.