Recently, a number of grasp detection methods have been proposed that can be used to localize robotic grasp configurations directly from sensor data without estimating object pose. The underlying idea is to treat grasp perception analogously to object detection in computer vision. These methods take as input a noisy and partially occluded RGBD image or point cloud and produce as output pose est...

Many of the tasks that are potential candidates for automation involve grasping. We are interested in the programming of robots to perform grasping tasks. To do this, we propose the notion of “perceptual programming,” where the key idea is to enable a system to observe a human performing a task, understand it, and perform the task with minimal human intervention. This allows the programmer to e...

We present in this paper the crucial components of “Mr.ArmHandOne” project. We shall discuss topics concerning the robot architecture; the robot building process, including the mechanics and electronics solutions to the robot structure; the description of a dynamic neural network for processing the raw data coming from the sonars; the local map dynamic construction, and the overall structure of...

When manipulating parts, it is important to determine the orientation of the part with respect to thegripper. This orientation may not be known precisely or may be disturbed by the act of grasping. It isin some cases possible to use mechanical compliance to orient parts during grasping. It was recentlyshown [14] that any part with polygonal boundary can be oriented and grasped i...

With the goal of advancing the state of automatic robotic grasping, we present a novel approach that combines machine learning techniques and physical validation on a robotic platform to develop a comprehensive grasp predictor. After collecting a large grasp sample set (522 grasps), we first conduct a statistical analysis of the predictive ability of grasp quality metrics that are commonly used...

Path-relinking is a major enhancement to the basic greedy randomized adaptive search procedure (GRASP), leading to significant improvements in solution time and quality. Path-relinking adds a memory mechanism to GRASP by providing an intensification strategy that explores trajectories connecting GRASP solutions and the best elite solutions previously produced during the search. This paper revie...

We distinguish between two types of grasp stability, which we call spatial grasp stability and contact grasp stability. We show via examples that spatial stability cannot capture certain intuitive concepts of grasp stability and hence that any full understanding of grasp stability must include contact stability. We derive a model of how the positions of the points of contact evolve in time on t...

Our goal is to automate the understanding of natural hand-object manipulation by developing computer visionbased techniques. Our hypothesis is that it is necessary to model the grasp types of hands and the attributes of manipulated objects in order to accurately recognize manipulation actions. Specifically, we focus on recognizing hand grasp types, object attributes and actions from a single im...

Understanding how the CNS controls reach-to-grasp will require behavioral and neurophysiological studies of reach-to-grasp in the monkey, including the evaluation of whole-hand grasp with explicit force requirements. In this study, monkeys performed a reach-to-grasp task in which the size, shape, and orientation of the objects were varied. The monkeys were required to grasp each object at five ...