Occlusion-Aware Search for Object Retrieval in Clutter

TL;DR

This paper presents a hybrid, occlusion-aware planning approach for robotic object retrieval in cluttered environments, combining learned distributions and reinforcement learning to efficiently locate hidden objects in real-time.

Contribution

It introduces a data-driven hybrid planner that predicts object locations and guides search actions using reinforcement learning, enabling effective retrieval in cluttered scenes.

Findings

Successful real-time retrieval in cluttered environments

Effective occlusion reasoning through learned distributions

Robust performance in both simulation and real-world tests

Abstract

We address the manipulation task of retrieving a target object from a cluttered shelf. When the target object is hidden, the robot must search through the clutter for retrieving it. Solving this task requires reasoning over the likely locations of the target object. It also requires physics reasoning over multi-object interactions and future occlusions. In this work, we present a data-driven hybrid planner for generating occlusion-aware actions in closed-loop. The hybrid planner explores likely locations of the occluded target object as predicted by a learned distribution from the observation stream. The search is guided by a heuristic trained with reinforcement learning to act on observations with occlusions. We evaluate our approach in different simulation and real-world settings (video available on https://youtu.be/dY7YQ3LUVQg). The results validate that our approach can search and retrieve a target object in near real time in the real world while only being trained in simulation.