Learning Hierarchical Interactive Multi-Object Search for Mobile Manipulation

TL;DR

This paper introduces HIMOS, a hierarchical reinforcement learning approach enabling robots to perform complex multi-object search tasks involving manipulation and navigation in unstructured environments, demonstrated through extensive simulation and real-world experiments.

Contribution

The paper presents a novel hierarchical RL framework that combines exploration, navigation, and manipulation skills using a semantic map memory for multi-object search in unstructured environments.

Findings

HIMOS effectively transfers to new environments without retraining.

The approach demonstrates robustness to unseen subpolicies and robot kinematic variations.

Successful real-world deployment shows practical applicability.

Abstract

Existing object-search approaches enable robots to search through free pathways, however, robots operating in unstructured human-centered environments frequently also have to manipulate the environment to their needs. In this work, we introduce a novel interactive multi-object search task in which a robot has to open doors to navigate rooms and search inside cabinets and drawers to find target objects. These new challenges require combining manipulation and navigation skills in unexplored environments. We present HIMOS, a hierarchical reinforcement learning approach that learns to compose exploration, navigation, and manipulation skills. To achieve this, we design an abstract high-level action space around a semantic map memory and leverage the explored environment as instance navigation points. We perform extensive experiments in simulation and the real world that demonstrate that, with accurate perception, the decision making of HIMOS effectively transfers to new environments in a zero-shot manner. It shows robustness to unseen subpolicies, failures in their execution, and different robot kinematics. These capabilities open the door to a wide range of downstream tasks across embodied AI and real-world use cases.