Value of Assistance for Grasping
This addresses the challenge of efficient sensing in robotic grasping for applications like collaborative robotics, though it appears incremental as it builds on existing probabilistic estimation methods.
The paper tackles the problem of a robot needing to decide which sensing action to take before grasping an object with uncertain pose, by introducing a novel Value of Assistance (VOA) measure to assess the expected benefit of observations on task completion, and evaluates it in simulated and real-world collaborative grasping settings.
In multiple realistic settings, a robot is tasked with grasping an object without knowing its exact pose and relies on a probabilistic estimation of the pose to decide how to attempt the grasp. We support settings in which it is possible to provide the robot with an observation of the object before a grasp is attempted but this possibility is limited and there is a need to decide which sensing action would be most beneficial. We support this decision by offering a novel Value of Assistance (VOA) measure for assessing the expected effect a specific observation will have on the robot's ability to complete its task. We evaluate our suggested measure in simulated and real-world collaborative grasping settings.