Closed-Loop Next-Best-View Planning for Target-Driven Grasping

TL;DR

This paper introduces a reactive, closed-loop next-best-view planning method for target-driven robotic grasping that improves efficiency and robustness by adaptively exploring occluded scene parts based on ongoing scene reconstruction.

Contribution

It presents a novel reactive planning approach that dynamically guides exploration and grasping in cluttered scenes using continuous scene updates.

Findings

Reduces grasp execution times compared to fixed camera placements

Maintains high grasp success rates with adaptive exploration

Handles scenarios where fixed baselines fail

Abstract

Picking a specific object from clutter is an essential component of many manipulation tasks. Partial observations often require the robot to collect additional views of the scene before attempting a grasp. This paper proposes a closed-loop next-best-view planner that drives exploration based on occluded object parts. By continuously predicting grasps from an up-to-date scene reconstruction, our policy can decide online to finalize a grasp execution or to adapt the robot's trajectory for further exploration. We show that our reactive approach decreases execution times without loss of grasp success rates compared to common camera placements and handles situations where the fixed baselines fail. Video and code are available at https://github.com/ethz-asl/active_grasp.