Soft Tactile Sensor Embedded Artificial Skin
MetadataShow full item record
When making contact with objects, we perceive them as warm or cold, rough or smooth, and hard or soft using multiple mechanoreceptors. Current robots and prosthesis lack the perception of touch that is vital for in-hand manipulation and finger-object interaction, thus struggling on certain tasks such as slip prevention, grip control, and texture/stiffness recognition. Tactile feedback on robot manipulators and prosthetic hands are important advancement because it enables manipulation in unstructured surroundings, reveals surface/volumetric properties of objects and improves robotic/prosthetic autonomy. Sensor skin can provide rich, real-time tactile information to aid manipulation and can conformally wrap around a variety of existing fingertips. Numerous soft tactile sensors have been developed using liquid metal (eutectic Gallium Indium, or eGaIn) and flexible elastomer. These sensor skins are inferior to human tactile sensing performance in terms of sensitivity, spatial and/or temporal resolution. Current approaches to measure shear force suffer from poor resolution and ambiguity. A highly sensitive sensor skin that accurately resolves contact force in three-dimension and senses vibration is needed for artificial manipulator to better interact with the environment and external objects. This dissertation describes the design and development of a soft tactile sensing skin that is conformable to existing robotic manipulators and provides dynamic tactile sensing of normal and shear force as well as vibration. A bioinspired shear force sensor is developed by measuring the asymmetric strain pattern of sensor skin when shear force is applied. However normal force would induce symmetric strain pattern, analytically proving that the sensor response is independent of normal force. A 2D solid mechanics steady finite element analysis is developed to evaluate the sensor performance and determine geometric parameters of the artificial skin and strain sensor that provide adequate sensitivity over the light touch shear force range. Static characterization experiments are conducted to produce the linear calibration between sensor response and shear force. This relation is matches analytical estimations as well as simulation predictions. The artificial sensor skin is further examined dynamically in stepwise unloading, slip and controlled vibration tests. We show that the sensor has potential of detecting insipient slip and can resolve vibrations equivalent, or better, than humans. The sensor resolves a variety of tactile events during pick and place, drop or handoff tasks on a robotic manipulator. The shear tactile sensor skin is extended to two-dimensions and integrated with a normal force sensor. The resistive normal force sensor is based on deformation of liquid metal filled spiral shaped microfluidic channel with respect to normal force. The normal force sensors exhibit sensitivity of 18 %/N and better-than-human performance to measure vibration. It is shown that the integrated sensor skin encodes spatially dispersed normal force and lumped shear force in two directions, although there are design and optimization challenges to match the sensitivity to one-dimension shear sensing skin. This research has resulted in the development of a flexible normal and shear sensing skin that is also capable of sensing vibration. The sensing skin can be applied to robotic manipulators or prosthetic hands to improve manipulation performance, prevent slip, gather surface/volumetric object properties for autonomous robot or smarter and more user-friendly prosthesis.
- Mechanical engineering 
Showing items related by title, author, creator and subject.
Occupational Fatigue Prediction for Entry-Level Construction Workers in Material Handling Activities Using Wearable Sensors Lee, WonilResearch on the measurement and prediction of occupational fatigue of construction workers using wearable sensors has been carried out using different types of sensor technologies and measurement variables. Previous studies ...
Powers, ThomasWe develop a framework to select a subset of sensors from a field in which the sensors have an ingrained independence structure. Given an arbitrary independence pattern, we construct a graph that denotes pairwise independence ...
Goel, MayankComputing devices continually evolve and in a relatively short time, have morphed from a room-sized machine to something much smaller. Today, a mobile computer can be kept in our pocket, or worn as a wristwatch. These ...