Vision-based grasp planning can be approached as an optimization problem, where a hand configuration that indicates a stable grasp needs to be located in a large search space. In this thesis, we proposed applying genetic algorithm (GA) to grasp planning of 3D object in arbitrary shapes and any robot hand. Details are given on the selection of operators and parameters of GA. GraspIt! simulator [2] is used for implementing the proposed algorithm and as the test environment. A quantitative analysis including the comparison with simple random algorithm and simulated annealing (SA) method is carried out to evaluate the performance of the GA based planner. Both GA and SA grasp planner are tested on different sets of hand-object. And two different quality metrics are used in the planning. Given the same amount of time, GA is shown to be capable of finding a force-closure grasp with higher stability than SA.

Grasping can be a complicated process for robotics due to the replication of human fine motor skills and typically high degrees of freedom in robotic hands. Robotic hands that are underactuated provide a method by which grasps can be executed without the onerous task of calculating every fingertip placement. The general shape configuration modes available to underactuated hands lend themselves well to an approach of grasping by shape primitives, and especially so when applied to gloveboxes in the nuclear domain due to the finite number of objects anticipated and the safe assumption that objects in the set are rigid. Thus, the object set found in a glovebox can be categorized as a small set of primitives such as cylinders, cubes, and bowls/hemispheres, etc. These same assumptions can also be leveraged for reliable identification and pose estimation within a glovebox. This effort develops and simulates a simple, but robust and effective grasp planning algorithm for a 7DOF industrial robot and three fingered dexterous, but underactuated robotic hand. The proposed grasping algorithm creates a grasp by generating a vector to the object from the base of the robot and manipulating that vector to be in a suitable starting location for a grasp. The grasp preshapes are selected to match shape primitives and are built-in to the Robotiq gripper used for algorithm demonstration purposes. If a grasp is found to be unsuitable via an inverse kinematics solution check, the algorithm procedurally generates additional grasps to try based on object geometry until a solution can be found or all possibilities are exhausted. The algorithm was tested and found capable of generating valid grasps for visually identified objects, and can recalculate grasps if one is found to be incompatible with the current kinematics of the robotic arm. / text

Robotic grasping and manipulation of three-dimensional deformable objects is a complex task that currently does not have robust and flexible solutions. Deformable objects include a wide variety of elastic and inelastic objects that change size and shape during manipulation. The development of adaptable methods for grasping and autonomously controlling the shape of three-dimensional deformable objects will benefit many commercial applications, including shaping parts for assembly in manufacturing, manipulating food for packaging and controlling tissues during robotic surgery. Controlling a deformable object to a desired shape requires first choosing contact points on the object's surface. Next, the robotic hand is positioned in the correct position and orientation to grasp and deform the object. After deformation, the object is assessed to evaluate the quality of the shape control procedure. In many cases, this process is completed without knowing the object's properties or behaviour before deformation. This work proposes and implements the framework for a robotic arm and hand system to control the shape of a previously unseen deformable object autonomously. Significant original contributions are made in developing an original algorithm to plan contact points on a three-dimensional object for grasping and shape control. This research uses a novel object representation to reduce the dimensionality of the deformable object manipulation problem. A path planning algorithm guides the robot arm to the optimal valid grasp pose to deform the object at the determined contact points. Additional contributions include developing a multi-view assessment strategy to determine the quality of the deformation towards the desired shape. The system completes the objectives using depth and colour images captured from a single point of view to locate and identify a previously unseen three-dimensional object within a robotic workspace. After estimating the unknown object's geometry, initial grasp contact points are planned to control the object to the desired shape. The grasp points are used to plan and execute a collision-free trajectory for the robot manipulator to place the robotic hand in the optimal position and orientation to grasp and deform the object. After the deformation is complete, the object is moved to a variety of assessment positions to determine the success of the shape control procedure. The system is validated experimentally on a variety of deformable three-dimensional objects.

This report presents a system for generating a stable, feasible, and reachable grasp of a polyhedral object. A set of contact points on the object is found that can result in a stable grasp; a feasible grasp is found in which the robot contacts the object at those contact points; and a path is constructed from the initial configuration of the robot to the stable, feasible final grasp configuration. The algorithm described in the report is designed for the Salisbury hand mounted on a Puma 560 arm, but a similar approach could be used to develop grasping systems for other robots.

Grasping is a useful ability that allows manipulators to restrain objects to a desired location or trajectory. Whole arm grasps are grasps that use the entire surface of the manipulator to apply contacts to an object. The problem of determining the shape of an object and planning a grasp for that object with a snake-like robot are considered in this work. Existing algorithms that attempt to allow robots to plan and perform whole arm grasps are lacking, they either use restrictive assumptions or have unrealistic demands in terms of required hardware. The work presented here allows even the most basic of robots to plan grasps on unknown objects whilst using a minimum of assumptions.The new developed Octograsp algorithm is a method of gaining information regarding the shape of the object to be grasped through tactile information alone. This contact information is processed using an inverse convex hull algorithm to build a model of the object's shape and position. The performance of the algorithms are examined using both simulations and experimental hardware, it is shown that accuracy errors as low as 3.1% can be obtained. The accuracy of the model depends upon factors such as the complexity of the object and the suitability of the robot. Manipulators consisting of a large number of small links with relaxed rotational constraints outperform other configurations. It is also shown that the accuracy can be improved by between 11% and 17% by contacting the object from multiple orientations, whilst also encircling from multiple positions can provide a very large improvement of between 56% to 86%. These methods allow even the coarse contact information provided by the experimental equipment to attain a model with an accuracy error of only 26%.A second novel algorithm is described that uses the information provided from the first algorithm to plan strong grasps over the desired object. The algorithm takes, on average, 25.1 seconds to plan the grasp. The mean strength of the planned grasps is 0.3816 using the wrench ball measure, this is firmly in the very good region. Several robotic configurations, as well as objects, are used to test the performance of the algorithm. The optimal parameters of the algorithm are investigated by using the results of 51030 different tests. It is again shown that robots that consist of a large number of small links and with high rotational ability perform the best.

In near future, robots are envisioned to work alongside humans in professional anddomestic environments without significant restructuring of workspace. Roboticsystems in such setups must be adept at observation, analysis and rational de-cision making. To coexist in an environment, humans and robots will need tointeract and cooperate for multiple tasks. A fundamental such task is the manip-ulation of large objects in work environments which requires cooperation betweenmultiple manipulating agents for load sharing. Collaborative manipulation hasbeen studied in the literature with the focus on multi-agent planning and controlstrategies. However, for a collaborative manipulation task, grasp planning alsoplays a pivotal role in cooperation and task completion.In this work, a novel approach is proposed for collaborative grasping and manipu-lation of large unknown objects. The manipulation task was defined as a sequenceof poses and expected external wrench acting on the target object. In a two-agentmanipulation task, the proposed approach selects a grasp for the second agentafter observing the grasp location of the first agent. The solution is computed ina way that it minimizes the grasp wrenches by load sharing between both agents.To verify the proposed methodology, an online system for human-robot manipu-lation of unknown objects was developed. The system utilized depth informationfrom a fixed Kinect sensor for perception and decision making for a human-robotcollaborative lift-up. Experiments with multiple objects substantiated that theproposed method results in an optimal load sharing despite limited informationand partial observability.

La prensión y manipulación de objetos se ha convertido en un área de gran interés en robótica, especialmente debido al desarrollo de dispositivos de prensión diestra como las manos antropomórficas, que incrementan la flexibilidad y verstilidad de los brazos robóticos, permitiendo así la prensión y manipulación de una gran variedad de objetos con un solo efector final. Esta tesis aborda varios problemas de planificación asociados a la prensión y manipulación de objetos discretos arbitrarios, esto es, objetos de forma arbitraria descritos mediante nubes de puntos o mallas poligonales. La obtención de una prensión con clausura de fuerza (force-closure) y de una prensión localmente óptima se realiza mediante procedimientos de búsqueda orientada basados en razonamientos geométricos en el espacio de prensiones. La medida de calidad de prensión utilizada es la mayor fuerza generalizada de perturbación que la prensión puede resistir, independientemente de la dirección de la perturbación. Sin embargo, las manos mecánicas y dispositivos de prensión reales difícilmente pueden asegurar que los dedos toquen el objeto justamente en los puntos de contacto calculados. Las regiones de contacto independiente (ICRs) se definen de forma tal que un dedo colocado en cada ICR asegura una prensión con clausura de fuerza; estas regiones otorgan robustez frente a errores en el posicionamiento de los dedos. Esta tesis presenta un algoritmo para obtener las ICRs con cualquier número de contactos con o sin fricción sobre la superficie de cualquier objeto tridimensional, asegurando también una calidad mínima controlada. La aproximación planteada genera las ICRs creciéndolas alrededor de los puntos de contacto de una prensión inicial apropiada, por ejemplo una prensión localmente óptima. Este método se extiende también para el cálculo de ICRs cuando varios contactos están fijados de antemano. El concepto de regiones de no prensión (NGRs) se introduce en este trabajo. Las NGRs se definen de forma tal que un dedo colocado en cada NGR siempre produce una prensión sin clausura de fuerza, independientemente de la posición exacta de cada dedo. Las ICRs y NGRs se utilizan para explorar de forma eficiente el espacio de prensiones. Este espacio es construido mediante un método de muestreo que provee muestras de prensiones con o sin clausura de fuerza, que luego se utilizan para calcular ICRs o NGRs respectivamente, y que luego sirven para etiquetar las configuraciones del espacio de prensiones. Se presenta también una secuencia de muestreo determinístico que permite una exploración incremental y uniforme del espacio de prensiones. La generación del espacio de prensiones se utiliza posteriormente para resolver el problema de reprensión (regrasping), esto es, la obtención de trayectorias de las puntas de los dedos sobre la superficie del objeto para cambiar de una prensión inicial a una final sin perder la condición de la clausura de fuerza. La tesis incluye ejemplos de aplicación para ilustrar el desempeño y la relevancia de los algoritmos planteados. / Object grasping and manipulation has become an area of great interest in robotics, specially due to the development of dexterous grasping devices like anthropomorphic hands that increase the flexibility and versatility of the robot arms, allowing the grasping and manipulation of a large variety of objects with a single end effector. This thesis tackles several planning problems associated with grasping and manipulation of arbitrary discrete objects, i.e. objects described with a cloud of points or a polygonal mesh. The computation of a force closure grasp and a locally optimal grasp is tackled using oriented search procedures based on geometric reasoning in the wrench space. The grasp quality measure considered is the largest perturbation wrench that the grasp can resist independently of the perturbation direction. However, real mechanical hands and grasping devices can hardly assure that the fingers will precisely touch the object at the computed contact points. Independent contact regions (ICRs) such that a finger contact in each ICR ensures a force closure grasp, provide robustness in front of finger positioning errors. This thesis presents an approach to compute ICRs with any number of frictionless or frictional contacts on the surface of any 3D object, assuring a controlled minimum grasp quality. The approach generates the ICRs by growing them around the contact points of a given appropriated starting grasp, like for instance a locally optimal grasp. The approach is also extended to compute the ICRs when several contacts are fixed beforehand. The notion of Non-Graspable Regions (NGRs) is introduced in this work, such that a finger contact in each NGR always produces a non-force closure grasp independently of the exact position of each finger. The ICRs and NGRs are used to efficiently explore the grasp space. The grasp space is constructed using a sampling method that provides samples of force closure or non force closure grasps used to compute ICRs or NGRs, respectively, which are used to label the configurations of the grasp space. An efficient deterministic sampling sequence is provided to allow a good incremental and uniform exploration of the grasp space. The generation of the grasp space is then applied to solve the regrasping problem, i.e. to obtain trajectories of the fingertips on the object surface in order to change from an initial to a final grasp without losing the force closure condition. Application examples are included to illustrate the relevance and performance of the proposed approaches.

This thesis investigates the questions of where to grasp and how to grasp a given object with an articulated robotic grasping device. To this end, aspects of grasp synthesis and hand motion planning and control are investigated. Grasp synthesis is the process of determining a palm pose with respect to the target object, as well as a hand joint configuration and/or grasp contact points such that a successful grasp execution is allowed. Existing methods tackling the grasp synthesis problem can be categorized in analytical and empirical approaches. Analytical approaches are based on geometric, kinematic and/or dynamic formulations, whereas empirical methods aim at mimicking human strategies.An overarching idea throughout this thesis is to circumvent the curse of dimensionality, which is inherent in high-dimensional planning problems, by incorporating empirical data in analytical approaches. To this end, tools from the field of constrained optimization are used (i) to synthesize grasp families based on available prototype grasps, (ii) to incorporate heuristics capturing human grasp strategies in the grasp synthesis process and (iii) to encode demonstrated grasp motions in primitive motion controllers.The first contribution is related to the computation and analysis of grasp families which are represented as Independent Contact Regions (ICR) on a target object’s surface. To this end, the well-known concept of the Grasp Wrench Space for a single grasp is extended to be applicable for a set of grasps. Applications of ICR include grasp qualification by capturing the robustness of a grasp to position inaccuracies and the visual guidance of a demonstrator in a teleoperating scenario. In the second main contribution of this thesis, it is shown how to reduce the grasp solution space during the synthesis process by accounting for human approach strategies. This is achieved by imposing appropriate constraints to a corresponding optimization problem. A third contribution in this dissertation is made to reactive motion planning. Here, primitive controllers are synthesized by estimating the free parameters of corresponding dynamical systems from multiple demonstrated trajectories. The approach is evaluated on an anthropomorphic robot hand/arm platform. Also, an extension to a Model Predictive Control (MPC) scheme is presented which allows to incorporate state constraints for auxiliary tasks such as obstacle avoidance.