Thesis (M.E.)--University of Canterbury, 2006. / Typescript (photocopy). Includes bibliographical references (leaves 99-101). Also available via the World Wide Web.

Researchers have made tremendous advances in robotic grasping in the past decades. On the hardware side, a lot of robot hand designs were proposed, covering a large spectrum of dexterity (from simple parallel grippers to anthropomorphic hands), actuation (from underactuated to fully actuated), and sensing capabilities (from only open/close states to tactile sensing). On the software side, grasping techniques also evolved significantly, from open-loop control, classical feedback control, to learning-based policies. However, most of the studies and applications follow the one-way paradigm that mechanical engineers/researchers design the hardware first and control/learning experts write the code to use the hand. In contrast, we aim to study the interplay between the mechanical and computational aspects in robotic grasping. We believe both sides are important but cannot solve grasping problems on their own, and both sides are highly connected by the laws of physics and should not be developed separately. We use the term "Mechanical Intelligence" to refer to the ability realized by mechanisms to appropriately respond to the external inputs, and we show that incorporating Mechanical Intelligence with Computational Intelligence is beneficial for grasping. The first part of this thesis is to derive hand underactuation mechanisms from grasp data. The mechanical coordination in robot hands, which is one type of Mechanical Intelligence, corresponds to the concept of dimensionality reduction in Machine Learning. However, the resulted low-dimensional manifolds need to be realizable using underactuated mechanisms. In this project, we first collect simulated grasp data without accounting for underactuation, apply a dimensionality reduction technique (we term it "Mechanically Realizable Manifolds") considering both pre-contact postural synergies and post-contact joint torque coordination, and finally build robot hands based on the resulted low-dimensional models. We also demonstrate a real-world application on a free-flying robot for the International Space Station. The second part is about proprioceptive grasping for unknown objects by taking advantage of hand compliance. Mechanical compliance is intrinsically connected to force/torque sensing and control. In this work, we proposed a series-elastic hand providing embodied compliance and proprioception, and an associated grasping policy using a network of proportional-integral controllers. We show that, without any prior model of the object and with only proprioceptive sensing, a robot hand can make stable grasps in a reactive fashion. The last part is about developing the Mechanical and Computational Intelligence jointly --- to co-optimize the mechanisms and control policies using deep Reinforcement Learning (RL). Traditional RL treats robot hardware as immutable and models it as part of the environment. In contrast, we move the robot hardware out of the environment, express its mechanics as auto-differentiable physics and connect it with the computational policy to create a unified policy (we term this method "Hardware as Policy"), which allows RL algorithms to back-propagate gradients w.r.t both hardware and computational parameters and optimize them in the same fashion. We present a mass-spring toy problem to illustrate this idea, and also a real-world design case of an underactuated hand. The three projects we present in this thesis are meaningful examples to demonstrate the interplay between the mechanical and computational aspects of robotic grasping. In the Conclusion part, we summarize some high-level philosophies and suggestions to integrate Mechanical and Computational Intelligence, as well as the high-level challenges that still exist when pushing this area forward.

This thesis discusses three projects that revolve around the central concept of the control of artificial hands. The first part of the thesis discusses the design of a museum exhibit for the South Florida Science Center that allows the public to control an i-limb Revolution prosthetic hand using electromyograph (EMG) sensors. A custom armature was designed to house the EMG sensors that are used to control the prosthesis. The top arm of the armature utilized a double rocker design for a greater range of motion which allows the display to accommodate arm sizes ranging from small children to large adults. This display became open to the public in March of 2019. The second part of the thesis describes a new concept for a simultaneous multi-object grasp using the Shadow hand robotic hand. This grasp is tested in an experiment that involves grasp and transportation tasks. This experiment also aims to analyze the benefit of soft robotic haptic feedback armband during the grasp and transportation tasks when a simulated break threshold is imposed on the objects. The usefulness of the haptic feedback was further tested with a guess the object task where the subjects had to determine which object was in the hand based solely off the armband. The new grasp synergy was deemed a success as all subjects were able to use the control method effectively with very little initial training. It was also found that the haptic feedback greatly aided in the successfully completing the transportation tasks. The human subjects were asked to rate the haptic feedback after each task, the overall rating for the helpfulness of the haptic feedback was rated as 4.6 out of 5. The final part of the thesis discusses an approach at gaining additional control signals for a dexterous artificial hand using a brain computer interface. This project seeks to investigate three neuromarkers for control which are: mu, xi and alpha. During analysis, the mu rhythm was not seen in our subject but alpha and xi were. Using deep learning approaches at classification, we were able to classify alpha and xi with at least a 90 percent accuracy. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2019. / FAU Electronic Theses and Dissertations Collection

As robots become more prevalent in our everyday lives, both in our workplaces and in our homes, it becomes increasingly likely that people who are not experts in robotics will be asked to interface with robotic devices. It is therefore important to develop robotic controls that are intuitive and easy for novices to use. Robotic hands, in particular, are very useful, but their high dimensionality makes creating intuitive human-machine interfaces for them complex. In this dissertation, we study the control of non-anthropomorphic robotic hands by non-roboticists in two contexts: collaborative manipulation and assistive robotics. In the field of collaborative manipulation, the human and the robot work side by side as independent agents. Teleoperation allows the human to assist the robot when autonomous grasping is not able to deal sufficiently well with corner cases or cannot operate fast enough. Using the teleoperator’s hand as an input device can provide an intuitive control method, but finding a mapping between a human hand and a non-anthropomorphic robot hand can be difficult, due to the hands’ dissimilar kinematics. In this dissertation, we seek to create a mapping between the human hand and a fully actuated, non-anthropomorphic robot hand that is intuitive enough to enable effective real-time teleoperation, even for novice users. We propose a low-dimensional and continuous teleoperation subspace which can be used as an intermediary for mapping between different hand pose spaces. We first propose the general concept of the subspace, its properties and the variables needed to map from the human hand to a robot hand. We then propose three ways to populate the teleoperation subspace mapping. Two of our mappings use a dataglove to harvest information about the user's hand. We define the mapping between joint space and teleoperation subspace with an empirical definition, which requires a person to define hand motions in an intuitive, hand-specific way, and with an algorithmic definition, which is kinematically independent, and uses objects to define the subspace. Our third mapping for the teleoperation subspace uses forearm electromyography (EMG) as a control input. Assistive orthotics is another area of robotics where human-machine interfaces are critical, since, in this field, the robot is attached to the hand of the human user. In this case, the goal is for the robot to assist the human with movements they would not otherwise be able to achieve. Orthotics can improve the quality of life of people who do not have full use of their hands. Human-machine interfaces for assistive hand orthotics that use EMG signals from the affected forearm as input are intuitive and repeated use can strengthen the muscles of the user's affected arm. In this dissertation, we seek to create an EMG based control for an orthotic device used by people who have had a stroke. We would like our control to enable functional motions when used in conjunction with a orthosis and to be robust to changes in the input signal. We propose a control for a wearable hand orthosis which uses an easy to don, commodity forearm EMG band. We develop an supervised algorithm to detect a user’s intent to open and close their hand, and pair this algorithm with a training protocol which makes our intent detection robust to changes in the input signal. We show that this algorithm, when used in conjunction with an orthosis over several weeks, can improve distal function in users. Additionally, we propose two semi-supervised intent detection algorithms designed to keep our control robust to changes in the input data while reducing the length and frequency of our training protocol.

Η παρούσα διδακτορική διατριβή πραγματεύεται την κατασκευή και τον έλεγχο ενός καινοτόμου τεχνητού χεριού, για προσθετικές κυρίως εφαρμογές, κάνοντας χρήση βιομιμητικών ενεργοποιητών και πιο συγκεκριμένα ειδικά κατεργασμένων λεπτών κυλινδρικών αγωγών από μορφομνήμονα μεταλλικά κράματα νικελίου-τιτανίου. Εκμεταλλευόμενοι τα συγκριτικά πλεονεκτήματα των ενεργοποιητών αυτών έναντι των αντίστοιχων συμβατικών, αναπτύσσεται μια πλήρως λειτουργική συσκευή με μικρό μέγεθος και βάρος, ανθρωπομορφική εμφάνιση, αθόρυβη λειτουργία και χαμηλό κόστος κατασκευής και συντήρησης, ικανή να εκπληρώσει σε μεγάλο βαθμό τις απαιτήσεις των ατόμων με αναπηρία στα άνω άκρα. Για τη φυσική υλοποίηση του σκελετού του τεχνητού αυτού χεριού χρησιμοποιείται η τεχνολογία της ταχείας προτυποποίησης. Καθένα από τα πέντε δάκτυλά του ελέγχεται ανεξάρτητα μέσω ενός υπο-ενεργοποιούμενου μηχανισμού κίνησης με τεχνητούς τένοντες. Για τον έλεγχο θέσης κάθε δακτύλου, αναπτύσσεται και εφαρμόζεται μία νέα μέθοδος ελέγχου βασισμένη στην έμφυτη δυνατότητα ανάδρασης θέσης των προαναφερθέντων ενεργοποιητών μέσω μέτρησης της ηλεκτρικής τους αντίστασης. Επιπλέον, αναπτύσσεται κατάλληλος αλγόριθμος για τον σχηματισμό διαφόρων θέσεων και συλλήψεων του τεχνητού χεριού. Για τη βελτίωση του ελέγχου, το χέρι εξοπλίζεται με αισθητήρες αφής στα ακροδάκτυλα, καθώς και με τη δυνατότητα οδήγησης συσκευών οπτικής και απτικής ανάδρασης. Όλα τα ηλεκτρονικά κυκλώματα που είναι απαραίτητα για την οδήγηση των ενεργοποιητών και τον έλεγχο του χεριού αναπτύσσονται και ενσωματώνονται στο εσωτερικό του φυσικού πρωτοτύπου. Με τη βοήθεια ειδικού προγραμματιστικού πακέτου, σχεδιάζεται μία γραφική διεπαφή ελέγχου μέσω της οποίας μελετάται και αξιολογείται η δυνατότητα του αναπτυχθέντος χεριού σε πειράματα σύλληψης διαφόρων αντικειμένων. Τέλος, προτείνονται διάφορες τεχνικές ελέγχου του χεριού από τους χρήστες του, ενώ αναπτύσσεται και κατάλληλος αλγόριθμος ελέγχου βασισμένος στη χρήση ηλεκτρομυογραφικών σημάτων. / This doctoral thesis presents the development and control of an innovative artificial hand, mostly for use in prosthetic applications, utilizing biomimetic actuators, and, more specifically, specially processed thin cylindrical wires made of shape memory nickel-titanium alloys. By exploiting the comparative advantages of these actuators over the conventional ones, a fully functional device is developed, of low size and weight, anthropomorphic appearance, silent operation, low fabrication and maintenance cost, which is capable of satisfying to a great extent the needs of the upper limb amputees. The physical implementation of the chassis of this artificial hand has been performed using rapid prototyping technology. Each of its five digits is independently controlled via a tendon-driven underactuated mechanism. For the position control of each digit, a novel control scheme is devised and implemented based on the inherent position feedback capability of these actuators via the measurement of their electrical resistance. In addition, the necessary algorithm is developed for the formation of various hand postures and prehension patterns. In order to improve the overall hand control, the hand is equipped with tactile sensors at its fingertips, and is also capable of driving optical and tactile feedback devices. All the necessary electronics for driving the actuators and controlling the hand are developed and embedded inside the physical prototype. Using a special programming package, a graphical user interface is designed, through which the grasp capabilities of the developed hand are studied and evaluated for various objects. Finally, several user control techniques of the hand are proposed, and a control algorithm based on the use of electromyographic signals is also developed.

Les travaux exposés dans cette thèse sont de natures multiples mais visent tous à une meilleure compréhension du geste de saisie chez l'homme, que ce soit d'un point de vue comportemental, cinématique ou de contrôle. Lorsqu'il doit saisir un objet, l'homme s'appuie sur une structure de contrôle multi-niveaux ainsi que son expérience, ce qui lui permet d'estimer les mouvements à effectuer de manière très efficace avant même d'avoir commencé à bouger. Nous pensons que ce mode de commande peut apporter une solution innovante au double problème de l'atteinte et de la saisie par une main artificielle. Nous avons donc développé une architecture de commande distribuée reproduisant en partie ces mécanismes et capable de contrôler ce genre d'artefacts de manière efficace, déclinée en plusieurs versions en fonction du niveau de contrôle souhaité. Elle est constituée d'un ensemble d'unités d'appariement s'inspirant des structures présentes dans le Système Nerveux Central : chacune a en charge une partie du problème global à résoudre, elles intègrent des informations en provenance de la consigne et/ou d'autres unités à travers des échanges parfois redondants, et elles s'appuient sur un algorithme d'apprentissage supervisé. Afin de mieux comprendre les principes qui sous-tendent le mouvement humain nous nous sommes aussi intéressés à la modélisation de la main et du geste de saisie, que ce soit à travers un protocole d'expérimentation chez l'homme ou l'analyse de données médicales et vidéos chez le singe / Works presented in this thesis are of multiple kinds but all aim at a better understanding of the human grasping movements, may it be from a behavioural, kinematics or control point of view. When one wants to grasp an object he relies on a multilayer control structure and its personal experience, the two of which allow him to estimate the appropriate move in a very efficient way, even before he actually started to move. We think that this type of command can bring forth an innovative solution to the double reach and grasp problem that face an artificial hand. We developed a distributed command architecture that reproduce in part these mechanisms and is able to control this type of artefacts in an efficient way, several versions of which were implemented regarding the desired control level. It consists of a group of matching units that takes inspiration in the Central Nervous System: each of them is in charge of a part of the global problem to be solved; they integrate data from the system inputs and/or from other units in partly redundant ways; and they rely on a supervised learning algorithm. In order to better understand the underlying principles of human movement we also took interest in hand and grasping movement models, may it be through an experimental protocol on human or monkey medical and video data analysis