No description available.

Persons with visual impairments often rely on navigational electronic aids, which typically employ speech commands for guidance through novel routes. However, navigational speech commands may interfere with the perception of acoustically rich environmental information, resulting in potentially detrimental effects. We investigated the sense of touch as a means to convey navigational commands instead. The somatotopic representation of the body surface within the central nervous system makes spatial information intuitive to our skin, suggesting that the tactile channel should be equivalent to, if not better than, the auditory channel at processing directional commands. Additionally, based on Wickens’ Multiple resource theory, the tactile channel should mitigate the sensory load in the auditory channel in travelers with visual impairments. We tested the ability of blind users to process directional commands conveyed via a tactile navigational belt. 14 blind participants were tested with the tactile belt under conditions of either low or high acoustic sensory load, simulating different outdoor environments. For comparison, the same participants were tested also with a conventional auditory device. Consistent with previous studies, we found navigation with the tactile belt to be less efficient than navigation with the auditory aid in the absence of environmental sounds. However, we found also – for the first time, to our knowledge – that tactile performance was less compromised under conditions of high acoustic sensory load. These results will help to inform the further investigation and development of tactile displays to benefit blind travelers. / Thesis / Master of Science (MSc)

The EU funded HAPTEX project aimed to create a virtual reality system that allowed a user to explore and manipulate a suspended virtual textile with the thumb and index finger. This was achieved through a combination of a tactile renderer on the fingertips for surface textures and a force feedback system for deformation of the virtual material. This project focuses on the tactile rendering component of this system, which uses a tactile display developed at the University of Exeter. The 24 pin display is driven by piezoelectric bimorphs. Each of the pins can be driven independently, allowing for a variety of different sensations to be transmitted to the fingertip. The display is driven by rendering software that uses a spatial spectrum of the intended surface, in combination with the frequency response of touch receptors in the skin, position on the surface, and exploration velocity to produce a signal that is intended to recreate the sensation of exploring the surface texture. The output signal on each of the 24 contactors is a combination of high (320 Hz) and low (40 Hz) frequency sine waves. In this project, the tactile renderer is initially evaluated based on its ability to recreate the sensations of exploring particular textured surfaces. The users were asked to rank virtual textures in order of similarity to a real target texture. The results of the initial test were disappointingly low, with a 38.1±3.1% correct identification rate. However, feedback from this initial test was used to make improvements to the rendering strategy. These improvements did not give a significant improvement in identification (41.3±1.6%). Finally, the tests were repeated with a target virtual texture instead of the real one used in previous tests. This test yielded a higher identification rate (64.1±5.5%). This increase in identification suggests that the virtual textures are distinguishable but that they not always accurate recreations of the real textures they are mimicking.

Des métastases cérébrales vont se développer chez 10 à 30% des patients atteints de cancer. La radiothérapie fait partie des possibilités de traitement, et ceci même si les dommages induits au cerveau par des rayonnements ionisants sont potentiellement importants. Nous proposons l’utilisation de l’Imagerie par Résonance Magnétique fonctionnelle (IRMf) sur le rat pour mieux comprendre ces effets. Ce mémoire traite de la mise en place d’un tel protocole d’IRMf. Les principaux points abordés sont la préparation de l’animal, les différentes insultes et stimulations sensorielles possibles ainsi que la méthode d’acquisition. Notre protocole d’insulte hyperoxique permet de déceler des dommages physiques d’origine vasculaire suite à une intense irradiation dans le cerveau du rat. Toutefois, la même procédure associée à une stimulation mécanique de la patte arrière de l’animal n’amène pas de changement observable à l’IRMf sur un sujet sain. Malgré tout, ce type de stimulation induit une réponse respiratoire, même sous anesthésie d’isoflurane. Une telle méthode n’est donc pas adéquate à l’étude d’animaux anesthésiés, surtout ceux dont la réponse cérébrale pourra avoir été réduite par une irradiation. Quelques améliorations et modifications du protocole seraient possiblement à même de permettre une mesure reproductible de la réponse d’IRMf à une stimulation sensorielle. Le présent mémoire décrit les tentatives de mise en place d’une stimulation sensorielle donnant lieu à une activation IRMf reproductible et localisée. De plus, un protocole de traitement d’image adapté au petit animal ainsi qu’une implémentation de la méthode keyhole ont été mis en place. L’insulte hyperoxique et ses effets sur le cerveau de rat ont été explorés plus en détail.

We consider the problem of enabling a robot to efficiently obtain a dense haptic map of its visible surroundings Using the complementary properties of vision and tactile sensing. Our approach assumes that visible surfaces that look similar to one another are likely to have similar haptic properties. In our previous work, we introduced an iterative algorithm that enabled a robot to infer dense haptic labels across visible surfaces in an RGB-D image when given a sequence of sparse haptic labels. In this work, we describe how dense conditional random fields (CRFs) can be applied to this same problem and present results from evaluating a dense CRF’s performance in simulated trials with idealized haptic labels. We evaluated our method using several publicly available RGB-D image datasets with indoor cluttered scenes pertinent to robot manipulation. In these simulated trials, the dense CRF substantially outperformed our previous algorithm by correctly assigning haptic labels to an average of 93% (versus 76% in our previous work) of all object pixels in an image given the highest number of contact points per object. Likewise, the dense CRF correctly assigned haptic labels to an average of 81% (versus 63% in our previous work) of all object pixels in an image given a low number of contact points per object. We compared the performance of dense CRF using uniform prior with a dense CRF using prior obtained from the visible scene using a Fully Convolutional Network trained for visual material recognition. The use of the convolutional network further improves the performance of the algorithm. We also performed experiments with the humanoid robot DARCI reaching in a cluttered foliage environment while using our algorithm to create a haptic map. The algorithm correctly assigned the label to 82.52% of the scenes with trunks and leaves after 10 reaches into the environment.

This dissertation describes the development of an assist-device aimed to deliver 3D graphic information to the visually impaired people. A human-in-loop approach was used to analyze whether a virtual 3D shape can be transferred correctly to the human users.The proposed device in this dissertation consists of two major parts: (a) A system of position sensors for real time localization based on magnetization, and (b) A single vibratory actuator working at varied frequencies based on its real time location. The error bound of the position measurement was tested to be 2 mm, which defined the machine resolution of the shape display. In order to realize the refresh rate of the localization that can follow user's scanning speed, the parallel data processing sequences for computer and microcontroller were designed. Additionally, vibratory electromagnetic (EM) actuators were discussed based on eddy current and permanent magnet methods. The simulation study showed that eddy current method was not applicable for millimeter size coil. Accordingly, the permanent magnet method was developed and the force detection threshold of human tactile perceptions was studied.Virtual shape perception experiments were made with participation of 3 volunteers who were not aware of the 3D shape information prior to the tests. Based on the four sets of shape tests, we conclude that the majority of the shape information is able to be delivered to users by using the proposed device. Difficulties for perceiving the local sharp profile e.g. thin plates and large curvature in small shapes may be better addressed by multiple actuators simultaneously providing shape information in the local boundary detection.The major contribution of this dissertation is the 3D shape display implemented by a miniature and low cost device. The developed device utilizes both passive stimulation and active search so that a commonly used large scale actuators matrix based on mere active touch method is avoided. The studies on the required force/energy input from the actuator showed that EM actuators can be miniaturized to millimeter scale without sacrificing the ability to induce tactile stimulation. Additional uniqueness of the proposed system is the ability to present hollow features, which is impossible to display by the existing devices.

Purpose - The purpose of this paper is to explore what people think is the best way to learn about innovations by different learning styles and characteristics. The paper will give answers to which learning style that people think is the best way to use when learning about innovation. Methodology - The data for this study were collected via online-surveys and through paper surveys. A total of 224 usable responses were obtained. The method will consist of primary data, which will be collected through surveys. Secondary research will be presented in the empire which will be retrieved from databases like Scopus, Diva, Emerald, Web of Science and Google Scholar. Implications/findings - Of the four different learning styles investigated in this study the result revealed that people best learn about innovations from the tactile/kinaesthetic learning style. The least pedagogic method to learn about innovations was the visual/verbal learning style. Paper type - Research paper

This thesis explores two novel ways of reducing the data complexity of tactile sensing. The thesis begins by examining the state-of-the art in tactile sensing, not only examining the sensor construction and interpretation of data but also the motivation for these designs. The thesis then proposes two methods for reducing the complexity of data in tactile sensing. The first is a low-power tactile sensing array exploiting a novel application of a pressure-sensitive material called quantum tunnelling composite. The properties of this material in this array form are shown to be beneficial in robotics. The electrical characteristics of the material are also explored. A bit-based structure for representing tactile data called Bitworld is then defined and its computational performance is characterised. It is shown that this bit-based structure outperforms floating-point arrays by orders of magnitude. This structure is then shown to allow high-resolution images to be produced by combining low resolution sensor arrays with equivalent functional performance to a floating-point array, but with the advantages of computational efficiency. Finally, an investigation into making Bitworld robust in the presence of positional noise is described with simulations to verify that such robustness can be achieved. Overall, the sensor and data structure described in this thesis allow simple, but effective tactile systems to be deployed in robotics without requiring a significant commitment of computational or power resources on the part of a robot designer.

Legged locomotion provides robots with the capability of adapting to different terrain conditions. General complex terrain traversal methodologies solely rely on proprioception which readily leads to instability under dynamical situations. Biological legged locomotion utilizes somatosensory feedback to sense the real-time interaction of the feet with ground to enhance stability. Nevertheless, limited attention has been given to sensing the feet-terrain interaction in robotics. This project introduces a paradigm shift in robotic walking called sensitive walking realized through the development of a compliant bipedal platform. Sensitive walking extends upon the success of sensitive manipulation which utilizes tactile feedback to localize an object to grasp, determine an appropriate manipulation configuration, and constantly adapts to maintain grasp stability. Based on the same concepts of sensitive manipulation, sensitive walking utilizes podotactile feedback to enhance real-time walking stability by effectively adapting to variations in the terrain. Adapting legged robotic platforms to sensitive walking is not as simple as attaching any tactile sensor to the feet of a robot. The sensors and the limbs need to have specific characteristics that support the implementation of the algorithms and allow the biped to safely come in contact with the terrain and detect the interaction forces. The challenges in handling the synergy of hardware and sensor design, and fabrication in a podotactile-based sensitive walking robot are addressed. The bipedal platform provides contact compliance through 12 series elastic actuators and contains 190 highly flexible tactile sensors capable of sensing forces at any incident angle. Sensitive walking algorithms are provided to handle multi-legged locomotion challenges including stairs and irregular terrain.

La dernière décennie a vu s'établir la démocratisation des interfaces tactiles. De nombreux logiciels jusque là réservés aux ordinateurs de bureau offrent désormais une version qui se contrôle du bout des doigts. Cependant, l'expressivité limitée de la modalité tactile restreint drastiquement le nombre de fonctionnalités disponibles. La recherche explore donc différentes pistes pour augmenter cette expressivité notamment par l'identification des doigts. Alors que la littérature se focalise principalement sur les méthodes d'identification des doigts, cette thèse vise à mieux en comprendre l'utilisation afin de guider la conception de techniques d'interaction.Nous réalisons d'abord une revue des technologies existantes, présentons nos prototypes et évaluons l'utilisation de la reconnaissance des empreintes digitales, qui nous semble la solution la plus prometteuse en vue d'une intégration dans des systèmes commerciaux. Par la réalisation d'expériences contrôlées, nous étudions ensuite les différences de performances et de préférences entre les doigts, l'influence de l'identification des doigts sur la stratégie de réalisation de tâches. Nous explorons également comment intégrer cette information au sein des techniques d'interaction existantes et aider les utilisateurs à appréhender le riche vocabulaire offert. / During the last decade, touch interfaces have become more and more ubiquitous. A lot of software applications initially designed for desktop computers have now a tactile counterpart. However, the limited expressiveness of the touch modality restricts drastically the amount of features available on touch applications. Researchers have been exploring different tracks on how to augment this expressiveness, notably through finger identification. While the literature mainly focuses on tackling the technological challenge, this dissertation aims at better understanding its use in order to provide guidelines for designing interaction techniques.We first summarize existing technologies, present our prototypes and evaluate the use of fingerprint recognition, that we perceive as the most promising solution with the aim of being integrated in consumer devices. Through controlled experiments, we study the differences of performance and preference between fingers, the influence of finger identification on user strategies to complete tasks. We also explore how to integrate this new information along existing interaction techniques and help users grasp the rich vocabulary provided. Our works lead to guidelines for designing interaction techniques leveraging finger identification that we implement in three different interaction contexts (tabletops, tablets and smartphones) through Adoiraccourcix, an interaction technique combining command selection and parameter manipulation.