Sensing Methods for Soft Robotics

Abstract

Soft robots exhibit complex behaviors that emerge from deliberate compliance in the actuators and structure. This compliance allows soft robots to passively conform to the constraints of their environment and to the objects they are manipulating. Many soft robots are actuated by the flexible expansion of hermetically sealed volumes. Systems based on these principles are lightweight, flexible and have low reflected inertia. This makes them inherently safe in physical human robot interaction. Moreover, the sealed actuators and flexible joints are well-suited to work in harsh environments where external contaminates could breach the dynamic seals of rotating or sliding shafts.

Accurate motion control remains a highly challenging task for soft robotic systems. Precise models of the actuation dynamics and environmental interactions are often unavailable. This renders open-loop control impossible, while closed-loop control suffers from a lack of suitable feedback. Conventional motion sensors, such as linear or rotary encoders, are difficult to adapt to robots that lack discrete mechanical joints. The rigid nature of these sensors runs contrary to the aspirational benefits of soft systems. Other proposed soft sensor solutions are still in their infancy and have only recently been used for motion-control of soft robots.

This dissertation explores the design and use of inductance-based sensors for the estimation and control of soft robotic systems. These sensors are low-cost, lightweight, easy-to-fabricate and well-suited for the conditions that soft systems can best exploit. The inquiry of this dissertation is conducted both theoretically and experimentally for Fiber-Reinforced Elastomeric Enclosures (including McKibben muscles) and bellows actuators. The sensing of each actuator type is explored through models, design analyses and experimental evaluations. The results demonstrate that inductance-based sensing is a promising technology for these otherwise difficult-to-measure actuators. By combining sensing and actuation into a single component, the ideas presented in this work provide a simple, compact and lightweight way to create and control motion in soft robotic systems. This will enable soft systems that can interactively engage with their environment and their human counterparts.