Knowledge-Augmented Dexterous Grasping with Incomplete Sensing

Humans can determine a proper strategy to grasp an object according to the measured physical attributes or the prior knowledge of the object. This paper proposes an approach to determining the strategy of dexterous grasping by using an anthropomorphic robotic hand simply based on a label or a description of an object. Object attributes are parsed from natural-language descriptions and augmented with an object knowledge base that is scraped from retailer websites. A novel metric named joint probability distance is defined to measure distance between object attributes. The probability distribution of grasp types for the given object is learned using a deep neural network which takes in object features as input. The action of the multi-fingered hand with redundant degrees of freedom (DoF) is controlled by a linear inverse-kinematics model of grasp topology and scales. The grasping strategy generated by the proposed approach is evaluated both by simulation and execution on a Sawyer robot with an AR10 robotic hand.