Recent progress in the development of artificial skin concepts is a result of the increased demand for providing environment perception such as touch and flow sensing to robots, prosthetics and surgical tools. Tactile sensors are the essential components of artificial skins and attracted considerable attention that led to the development of different technologies for mimicking the complex sense of touch in humans. This dissertation work is devoted to the development of a bioinspired tactile sensing technology that imitates the extremely sensitive hair-like cilia receptors found in nature. The artificial cilia are fabricated from permanent magnetic, biocompatible and highly elastic nanocomposite material, and integrated on a giant magneto-impedance magnetic sensor to measure the stray field. A force that bends the cilia changes the stray field and is therefore detected with the magnetic sensor, providing high performance in terms of sensitivity, power consumption and versatility. The nanocomposite is made of Fe nanowires (NWs) incorporated into polydimethylsiloxane (PDMS). Fe NWs have a high remanent magnetization, due the shape anisotropy; thus, they are acting as permanent nano-magnets. This allows remote device operation and avoids the need for a magnetic field to magnetize the NWs, benefiting miniaturization and the possible range of applications. The magnetic properties of the nanocomposite can be easily tuned by modifying the NWs concentration or by aligning the NWs to define a magnetic anisotropy. Tactile sensors are realized on flexible and rigid substrates that can detect flow, vertical and shear forces statically and dynamically, with a high resolution and wide operating range. The advantage to operate the sensors in liquids and air has been utilized to measure flows in different fluids in a microfluidic channel. Various dynamic studies were conducted with the tactile sensor demonstrating the detection of moving objects or the texture of objects. Overall, the results confirm the possibility to easily control the sensors’ performance with the cilia arrangement and dimensions. The cost effective mold-based microfabrication process and magnetic operation enable a high degree of integration, which together with the extremely low power consumption make the artificial cilia sensor reported in this dissertation an attractive solution for many applications.

American Cancer Society estimates that in 2021 nearly 300,000 women in the United States will be diagnosed with invasive breast cancer, and about 43,600 women will die from breast cancer. While many have access to health care and cancer screening, women from rural or underdeveloped communities often have limited access. Therefore, there is a need for an inexpensive and easy-to-use breast cancer identification device, which can be employed in small clinics to provide support to primary care physicians. This work aims to develop a method to characterize breast tumors and tissue using non-invasive imaging modalities. The proposed bimodal imaging system has tactile and multispectral imaging capabilities. Tactile imaging modality characterizes tumors by esti-mating their depth, size, and stiffness, along with the Tactile Index. Multispectral imaging modality identifies breast asymmetry, texture, and inflammation changes, together with the Spectral Index. These indices are combined with the BCRAT Index, the risk score devel¬oped by the National Institute of Health, to form the Multimodal Index for personalized breast cancer risk assessment. In this study, we will describe the development of the bimodal imaging system. We will present the algorithms for tactile and multispectral modalities. Tactile and Multispec¬tral Profile Diagrams are developed to capture broad imaging signals in a compact and application-specific way. A Tactile Profile Diagram is a pictorial representation of the rel¬ative depth, size, and stiffness of the imaged tumor. A Multispectral Profile Diagram is a representative pattern image for breast tissue superficial optical properties. To classify the profile diagrams, we employ the Convolutional Neural Network deep learning method. We will describe the results of the experiments conducted using tissue-mimicking phan¬toms and human in-vivo experiments. The results demonstrate the ability of the method to classify and quantify tumor and tissue characteristics. Finally, we describe the method to calculate Multimodal Index for the malignancy risk assessment via tactile and multispectral imaging modalities and the risk probability based on the health records. / Electrical and Computer Engineering

Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2007. / An abnormal sized mass in the neck is a common clinical finding and it can be the result of inflammation caused by bacterial or viral infection or it can be due to more serious diseases and malignant tumours. The most popular method of examining the neck is by manual palpation. Other methods include ultrasound, CT scan, MRI and PET. These methods though are expensive to perform and require specialists to interpret the results. The aim of this thesis was to design and develop a neck palpation device for telemedicine applications. The device uses an array of Force Sensing Resistors (FSRs) attached to an inflatable bladder. The bladder is mounted to the inside of a neck brace and it is inflated with an air pump controlled by a computer. As the bladder inflates the sensors press against the patient’s neck and the necessary data can be collected. A technique known as image registration is used to improve the resolution of the images sensed with the FSRs. The device provides a reproducible record of the examination for both the surgeon and the patient’s medical record, and provides the patient information as if the doctor examined the patient with his own hands without physically being there. A prototype of the device was built and used to perform numerous tests. The tests were conducted using different objects which are inserted into a silicone neck to simulate different lymph nodes. The device was used to test for shape, smallest size, different sizes, repeatability and hardness. The results showed that the device works well for spherical objects of different sizes but gives unsatisfactory results when the objects have sharp edges and complex forms. The image registration algorithm enhanced the images to a good representation of the object. Different sizes could be distinguished as well as hardness to some extend.

This research addresses the ability of dissipative passive actuators to generate control effects on a passive haptic interface. A haptic display is a human-machine interface that constructs a sensation of touch for the human operator. Applications can be found in various industries, space, medicine and construction etc. A dissipative passive haptic display contains passive actuators that can remove energy from the system by resisting motions in the system. The advantage of a dissipative passive haptic display is better safety compared to an active display. Its disadvantage is the limited control ability from the passive actuators. This research starts with the identification of the control ability and limitations of dissipative passive haptic interfaces. The ability is identified as the steerability, the ability to redirect motions of a manipulator. The force generation analysis of each individual actuator is then selected as an approach to evaluate the steerability. Steerability metrics are defined to evaluate the steerability. Even though non-redundant manipulators dont have desired steerability, optimal steering configurations are found for the best operation. Steerability is improved by redundancy in serial or parallel structures. A theorem is developed to evaluate steerability for redundant manipulators. The influence of system dynamics on their steerabilities is discussed. Previously developed haptic interfaces are evaluated based on their steerabilities. Steerability analysis of three-dimensional haptic interfaces is also given to a limited extent as an extension of the two-dimensional cases. Brakes and clutches are the two types of dissipative passive actuators in this research.

In this thesis, we propose a framework for automated assessment of surgical skills to expedite the manual assessment process and to provide unbiased evaluations with possible dexterity feedback. Evaluation of surgical skills is an important aspect in training of medical students. Current practices rely on manual evaluations from faculty and residents and are time consuming. Proposed solutions in literature involve retrospective evaluations such as watching the offline videos. It requires precious time and attention of expert surgeons and may vary from one surgeon to another. With recent advancements in computer vision and machine learning techniques, the retrospective video evaluation can be best delegated to the computer algorithms. Skill assessment is a challenging task requiring expert domain knowledge that may be difficult to translate into algorithms. To emulate this human observation process, an appropriate data collection mechanism is required to track motion of the surgeon's hand in an unrestricted manner. In addition, it is essential to identify skill defining motion dynamics and skill relevant hand locations. This Ph.D. research aims to address the limitations of manual skill assessment by developing an automated motion analysis framework. Specifically, we propose (1) to design and implement quantitative features to capture fine motion details from surgical video data, (2) to identify and test the efficacy of a core subset of features in classifying the surgical students into different expertise levels, (3) to derive absolute skill scores using regression methods and (4) to perform dexterity analysis using motion data from different hand locations.

Thesis (Ph. D.)--University of Missouri-Columbia, 2007. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on March 6, 2008) Includes bibliographical references.

Thesis (M. S.)--Mechanical Engineering, Georgia Institute of Technology, 2010. / Committee Chair: Book, Wayne; Committee Member: Glezer, Ari; Committee Member: Sadegh, Nader. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Somatosensory neurons densely innervate skin, our largest sensory organ. Adult skin continually remodels throughout the lifespan to maintain a protective barrier for our bodies. How sensory neurons maintain their peripheral endings in the face of continual turnover of their target tissue is not well understood. To address this gap in knowledge, I analyzed the temporal dynamics and mechanisms of structural plasticity of touch receptors in healthy adult skin. My studies focused on the terminals of Merkel-cell afferents in mouse touch domes. These two-part touch receptors comprise epithelial Merkel cells innervated by branching axons of fast-conducting sensory neurons. I show that Merkel cells and their afferents are structurally plastic over the course of hair growth in adults. These two components simplify during active hair growth, with fewer terminal neurites and fewer Merkel cells per touch dome at this stage compared with other phases of hair growth. Merkel-cell removal was observed with multiple molecular markers. Additionally, mice showed diminished touch-evoked behavior during hair growth compared with follicle quiescence. Next, I showed that Sarm1, a key effector of Wallerian degeneration, is not required for structural plasticity of Merkel cell-neurite complexes in young adulthood. Finally, I developed a technique to perform time-lapse in vivo imaging of identified Merkel cells and afferent terminals over the course of a month. These structures were highly plastic, with afferent terminals undergoing frequent growth and regression, as well as both Merkel cells and terminal branches being added or removed. Together, these studies reveal that peripheral nerve terminals undergo a previously unsuspected amount of structural plasticity in healthy tissue.

Robotic grasping and manipulation remains an active area of research despite significant progress over the past decades. Many existing solutions still struggle to robustly handle difficult situations that a robot might encounter even in non-contrived settings.For example, grasping systems struggle when the object is not centrally located in the robot's workspace. Also, grasping in dynamic environments presents a unique set of challenges. A stable and feasible grasp can become infeasible as the object moves; this problem becomes pronounced when there are obstacles in the scene. This research is inspired by the observation that object-manipulation tasks like grasping, pick-and-place or insertion require different forms of awareness. These include reachability awareness -- being aware of regions that can be reached without self-collision or collision with surrounding objects; tactile awareness-- ability to feel and grasp objects just tight enough to prevent slippage or crushing the objects; and 3D awareness -- ability to perceive size and depth in ways that makes object manipulation possible. Humans use these capabilities to achieve a high level of coordination needed for object manipulation. In this work, we develop techniques that equip robots with similar sensitivities towards realizing a reliable and capable home-assistant robot. In this thesis we demonstrate the importance of reasoning about the robot's workspace to enable grasping systems handle more difficult settings such as picking up moving objects while avoiding surrounding obstacles. Our method encodes the notion of reachability and uses it to generate not just stable grasps but ones that are also achievable by the robot. This reachability-aware formulation effectively expands the useable workspace of the robot enabling the robot to pick up objects from difficult-to-reach locations. While recent vision-based grasping systems work reliably well achieving pickup success rate higher than 90\% in cluttered scenes, failure cases due to calibration error, slippage and occlusion were challenging. To address this, we develop a closed-loop tactile-based improvement that uses additional tactile sensing to deal with self-occlusion (a limitation of vision-based system) and adaptively tighten the robot's grip on the object-- making the grasping system tactile-aware and more reliable. This can be used as an add-on to existing grasping systems. This adaptive tactile-based approach demonstrates the effectiveness of closed-loop feedback in the final phase of the grasping process. To achieve closed-loop manipulation all through the manipulation process, we study the value of multi-view camera systems to improve learning-based manipulation systems. Using a multi-view Q-learning formulation, we develop a learned closed-loop manipulation algorithm for precise manipulation tasks that integrates inputs from multiple static RGB cameras to overcome self-occlusion and improve 3D understanding. To conclude, we discuss some opportunities/ directions for future work.

Aujourd'hui, les robots sont de plus en plus présents dans la vie quotidienne. L’étude et le développement de stratégies d'interaction sociale et émotionnelle constitue un point clé de leur insertion dans notre espace social. Ces derniers années, beaucoup de recherches se sont intéressées à la communication homme-robot en exploitant les expressions faciales, posturales ou encore vocales, mais très peu de recherches se sont intéressées à l’interaction physique via le toucher. Cependant, des recherches récentes dans le domaine de la psychologie et des interfaces homme-machine (IHM) ont montré le rôle de la modalité haptique et plus particulièrement tactile dans la perception des émotions et de leurs différentes dimensions (par exemple valence, activation, dominance). L’objectif de ce projet est d’exploiter cette modalité sensorielle dans l'interaction affective homme-robot. Sur la base du robot humanoïde MEKA, un ensemble de capteurs tactiles et physiologiques seront étudiés et développés afin de sensibiliser certaines régions de son corps (ex. bras, épaule, main) et détecter l’état émotionnel de l’utilisateur. Par la suite, une série d’études seront menées afin d'analyser le comportement des utilisateurs dans des situations d’interaction affective avec le robot. Les résultats de ces études nous permettront d’identifier des comportements affectifs haptiques types qui seront utilisés pour modéliser le comportement du robot dans des contextes d’interactions sociales. / Today, robots are more and more present in everyday life. The study and the development of strategies of social and emotional interaction constitutes a key point of their insertion in our social space. The latter years, many researches were carried out in the man-robot communication by exploiting the facial expressions, posturals or still vocal, but very few focused on the physical interaction via the touch. However, recent researches in the field of the psychology and the human-machine interfaces (HMI) showed the role of the haptic modality and more particularly tactile in the perception of the feelings and their various dimensions (for example valence, activation, dominance). The objective of this project is to exploit this sensory modality in the emotional man-robot interaction. On the basis of the robot humanoid MEKA, a set of tactile and physiological sensors will be studied and developed to make sensitive certain regions of its body (eg arm, shoulder, hand) and to detect the emotional state of the user. Afterward, a series of studies will be led to analyze the behavior of the users in situations of emotional interaction with the robot. The results of these studies will allow us to identify typical haptic emotional behavior which will be used to model the behavior of the robot in contexts of social interactions.