Stable grasping with a multi-fingered robot hand: a behavior-based approach

The paper describes the software design approach and implementation of a stable grasp strategy to control a multifingered robot hand in a dynamic and uncertain environment. The overall design starts with a reactive system, called the Grasp Reactor, which measures the environment, and produces actions based on the environmental situations present. To improve the robustness of the Grasp Reactor, it is augmented with a deliberative component which executes concurrently with the Grasp Reactor. This component, called the Grasp Advisor, communicates global constraints to the Grasp Reactor to improve its decision making capability.