Multi-Resolution POMDP Planning for Multi-Object Search in 3D

05/06/2020
by   Kaiyu Zheng, et al.
0

Robots operating in household environments must find objects on shelves, under tables, and in cupboards. Previous work often formulate the object search problem as a POMDP (Partially Observable Markov Decision Process), yet constrain the search space in 2D. We propose a new approach that enables the robot to efficiently search for objects in 3D, taking occlusions into account. We model the problem as an object-oriented POMDP, where the robot receives a volumetric observation from a viewing frustum and must produce a policy to efficiently search for objects. To address the challenge of large state and observation spaces, we first propose a per-voxel observation model which drastically reduces the observation size necessary for planning. Then, we present an octree-based belief representation which captures beliefs at different resolutions and supports efficient exact belief update. Finally, we design an online multi-resolution planning algorithm that leverages the resolution layers in the octree structure as levels of abstractions to the original POMDP problem. Our evaluation in a simulated 3D domain shows that, as the problem scales, our approach significantly outperforms baselines without resolution hierarchy by 25 practicality of our approach on a torso-actuated mobile robot searching for objects in areas of a cluttered lab environment where objects appear on surfaces at different heights.

READ FULL TEXT

page 1

page 9

research
12/04/2020

Spatial Language Understanding for Object Search in Partially Observed Cityscale Environments

We present a system that enables robots to interpret spatial language as...
research
01/05/2021

Improving Training Result of Partially Observable Markov Decision Process by Filtering Beliefs

In this study I proposed a filtering beliefs method for improving perfor...
research
09/13/2023

Language-Conditioned Observation Models for Visual Object Search

Object search is a challenging task because when given complex language ...
research
10/19/2021

Towards Optimal Correlational Object Search

In realistic applications of object search, robots will need to locate t...
research
03/06/2023

A System for Generalized 3D Multi-Object Search

Searching for objects is a fundamental skill for robots. As such, we exp...
research
06/12/2019

Neural Graph Evolution: Towards Efficient Automatic Robot Design

Despite the recent successes in robotic locomotion control, the design o...
research
06/18/2020

Semantic Linking Maps for Active Visual Object Search

We aim for mobile robots to function in a variety of common human enviro...

Please sign up or login with your details

Forgot password? Click here to reset