UC San Diego

Rigid robotic systems struggle to solve problems that require drastic shape changes, possess high environmental uncertainty, and involve direct interaction with humans. The present work investigates four different approaches to designing mechanically reprogrammable robotic systems. The mechanical properties of the four presented systems are controlled by re-programming their geometry, structural properties, or dynamic properties during operation. The first study explores the potential of vibration-based locomotion in a self-folded robotic system to characterize and achieve controllable locomotion that is easily incorporated into laminate structures. We describe the extension of two-dimensional bristle-bot models to a three-dimensional model that explores parameters that govern linear and angular velocity, and implement a self-folding laminate-manufactured bristle-bot robot. In the second study, we present a new method for achieving controllable adhesion by vibrating a flexible plate near a surface, which generates a strong and controllable normal attraction force while allowing free motion parallel to the surface. Spatial pressure measurements demonstrate that adhesion is localized to the center of the disc. We developed a mobile robot capable of robustly acquiring payloads and adhering to vertical and inverted surfaces. In the third study, we present a variable stiffness surgical retractor that can be folded and inserted into a patient through a small opening, can change to a stiff state and anchor atraumatically to the body to retract an organ, and can be subsequently refolded and removed from the body. We model the variable stiffness mechanism (layer jamming) to predict the maximum load a retractor can support and compare this to experiments. In the fourth study, we present a model to predict the length scales at which jamming contributes to the overall performance of an inflatable beam, serving as a limb of a hybrid hard/soft robot. We measure the performance of prototype jamming limbs in cantilever loading tests, and experimentally test a quadrupedal system, using experiments to inform the design process. Incorporating jamming reinforcement into inflatable beams provides improved stiffness and graceful failure rather than the catastrophic buckling that normally characterizes inflated cylindrical beams. Together, these studies illuminate the exciting potential of mechanically reprogrammable robotic systems composed of laminate structures.