GRASP (programmation) | EPFL Graph Search

General responsibility assignment software patterns (ou principles), abrégé en GRASP, se composent de lignes directrices pour l'attribution de la responsabilité des classes et des objets en conception orientée objet. Les différents modèles et principes utilisés par les GRASP sont : le contrôleur, le créateur, l'indirection, le spécialiste de l'information, la cohésion forte, le couplage faible, le polymorphisme, les variations protégées, et l'invention pure. Tous ces modèles répondent à certains problèmes récurrents du développement logiciel. Ils n'ont pas été conçus pour créer une nouvelle façon de travailler, mais pour mieux documenter et normaliser l'existant à l'aide de principes de programmation éprouvés en conception orientée objet. D'après Craig Larman, . Ainsi, GRASP est surtout une boîte à outils mentale, une aide à la conception de logiciels orientés objets. Modèles Contrôleur Le contrôleur de schéma attribue la responsabilité de traiter les événemen

Enhancing our understanding of human fine manipulation skills and advancing robot dexterity in grasping

Kunpeng Yao

From surgery to watchmaking, fine-manipulation skills highly rely on the dexterity afforded by both hands. Coordination is key to human dexterity. Specifically, humans need not only to govern the abundant intrinsic degrees of freedom (DOFs) to allocate controls of task-demanded variables, but also to adapt postures in response to extrinsic task conditions. In spite of the recent advances in robotics research, human dexterity remains unattainable for robots, especially in terms of flexibility and adaptability. Therefore, it is necessary to gain insights into human fine-manipulation skills, to advance the dexterity of robots in similar tasks. This thesis deepens our understanding of human dexterity by investigating the human coordination in fine-manipulation skills taken from watchmaking craftsmanship.The first part of this thesis investigates both intrinsic and extrinsic coordination of upper-limbs in a non-symmetrical and non-rhythmic bimanual fine-manipulation task. We conduct a comparative study in the assembly of a spring on the watch face. Analysis of motion kinematics reveals that professional subjects mitigate the task challenge by actively modifying task conditions. Moreover, we offer a novel perspective to understand the skill acquisition in humans. We hypothesize that the evolution of coordination during skill improvement is driven by the changes in the structure of the optimal criterion of the central nervous system. We employ a bi-level optimization framework to infer the structure of this optimal criterion.The second part investigates how roles and control variables are distributed across hands and fingers. We compare task performance of human subjects under two experimental conditions when dismounting a screw from a watch face. When the watch face needed positioning, the role distribution of both hands was strongly influenced by hand dominance; when the watch face was stationary, a variety of hand pose combinations emerged. We propose a taxonomy of bimanual hand pose combinations and develop a graphical matrix-based representation approach to afford analysis of experimental observations. Our analysis suggests that the control of independent task demands is distributed across either hands or functional groups of fingers.In the third part, we take inspiration from human coordination principle to advance the dexterity of a robotic hand by exploiting its redundant DOFs. We propose a human-like algorithm to enable a robotic hand to grasp objects using arbitrary surface regions, no longer restricted to regular grasp types, such as pinch or power grasp. We present an iterative process to empower the robotic hand to grasp multiple objects in sequence. Moreover, we formulate a strategy to facilitate the exploitation of redundant DOFs for multitask planning. Our approaches have been validated both in simulation and on a real robotic hand.In summary, this thesis not only offers a deeper understanding of human dexterity by revealing both the intrinsic and extrinsic coordination of upper-limbs and the role distribution across hands in bimanual fine-manipulation tasks, but also proposes algorithms that enable multi-fingered robotic hands to achieve human-like dexterous grasping of a single or even multiple objects using arbitrary surface regions. This thesis offers the prospect of developing algorithms for robotic dexterous grasping and manipulation, and also inspires the design of novel robotic hands and manipulators.

EPFL2022

Learning Externally Modulated Dynamical Systems

Aude Billard, Klas Jonas Alfred Kronander, Nicolas Sommer

Dynamical Systems (DS) are often used to represent motion, with the advantage of being easy to learn from demonstrations. We present a method to modulate DS depending on an external signal, extending our previous work on Locally Modulated DS (LMDS, Kronander 2017). We present two applications of our system, which would not have been possible to achieve without taking external sensing into account in the DS motion formulation. The first application is a task of localization and grasping of objects, using our previous work on compliant tactile exploration. We successfully localize and grasp objects whose position is unknown, using touch in a simulated environment. In the second application, we teach a robot how to react to collisions in order to navigate between obstacles while reaching.

2017

Dynamic Grasp Adaptation

Miao Li

The human hand is an amazing tool, demonstrated by its incredible motor capability and remarkable sense of touch. To enable robots to work in a human-centric environment, it is desirable to endow robotic hands with human-like capabilities for grasping and object manipulation. However, due to its inherent complexity and inevitable model uncertainty, robotic grasping and manipulation remains a challenge. This thesis focuses on grasp adaptation in the face of model and sensing uncertainties: Given an object whose properties are not known with certainty (e.g., shape, weight and external perturbation), and a multifingered robotic hand, we aim at determining where to put the fingers and how the fingers should adaptively interact with the object using tactile sensing, in order to achieve either a stable grasp or a desired dynamic behaviour. A central idea in this thesis is the object-centric dynamics: namely, that we express all control constraints into an object-centric representation. This simplifies computa- tion and makes the control versatile to the type of hands. This is an essential feature that distinguishes our work from other robust grasping work in the literature, where generating a static stable grasp for a given hand is usually the primary goal. In this thesis, grasp adaptation is a dynamic process that flexibly adapts the grasp to fit some purpose from the objectâs perspective, in the presence of a variety of uncertainties and/or perturbations. When building a grasp adaptation for a given situation, there are two key problems that must be addressed: 1) the problem of choosing an initial grasp that is suitable for future adaptation, and more importantly 2) the problem of design- ing an adaptation strategy that can react adequately to achieve desired behaviour of the grasped object. To address challenge 1 (planning a grasp under shape uncertainty), we propose an approach to parameterizing the uncertainty in object shape using Gaussian Processes (GPs) and incorporate it as a constraint into contact-level grasp planning. To realize the planned contacts using different hands interchangeably, we further develop a prob- abilistic model to predict the feasible hand configurations, including hand pose and finger joints, given the desired contact points only. The model is built using the con- cept of Virtual Frame(VF), and it is independent from the choice of hand frame and object frame. The performance of the proposed approach is validated on two differ- ent robotic hands, an industrial gripper (4 DOF Barrett hand) and a humanoid hand (16 DOF Allegro hand) to manipulate objects of daily use with complex geometry and various texture (a spray bottle, a tea caddy, a jug and a bunny toy). In the second part of this thesis, we propose an approach to the design of adapta- tion strategy to ensure grasp stability in the presence of physical uncertainties of objects(object weight, friction at contacts and external perturbation). Based on an object-level impedance controller, we first design a grasp stability estimator in the object frame using the grasp experience and tactile sensing. Once a grasp is predicted to be unstable during online execution, the grasp adaptation strategy is triggered to improve the grasp stability, by either changing the stiffness at finger level or relocating the position of one fingertip to a better area.