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

  title={Multi-Resolution POMDP Planning for Multi-Object Search in 3D},
  author={Kaiyu Zheng and Yoonchang Sung and George Dimitri Konidaris and Stefanie Tellex},
  journal={2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
Robots operating in households must find objects on shelves, under tables, and in cupboards. In such environments, it is crucial to search efficiently at 3D scale while coping with limited field of view and the complexity of searching for multiple objects. Principled approaches to object search frequently use Partially Observable Markov Decision Process (POMDP) as the underlying framework for computing search strategies, but constrain the search space in 2D. In this paper, we present a POMDP… 

Figures from this paper

Towards Optimal Correlational Object Search
The Correlational Object Search POMDP is introduced, which models correlations while preserving optimal solutions with a reduced state space and a hierarchical planning algorithm is proposed to scale up COS-POMDPs for practical domains.
Planning Multimodal Exploratory Actions for Online Robot Attribute Learning
An algorithm called information-theoretic reward shaping (ITRS) is developed that actively addresses the trade-off between exploration and exploitation in On-RAL problems and demonstrates ITRS’ superiority in learning efficiency and identification accuracy.
Dialogue Object Search
A new task, dialogue object search, is introduced, where a robot is tasked to search for a target object in a human environment while engaging in a “video call” with a remote human who has additional but inexact knowledge about the target’s location.
ESNI: Domestic Robots Design for Elderly and Disabled People
A design framework: ESNI(Exploration, Segmentation, Navigation, Instruction) that apply to mobile agent, achieve some functionalities such as processing human commands, picking up a specified object, and moving an object to another location.
Spatial Language Understanding for Object Search in Partially Observed City-scale Environments
A convolutional neural network model is proposed that learns to predict the language provider's relative frame of reference (FoR) given environment context and achieves faster search and higher success rate compared to a keyword-based baseline without spatial preposition understanding.


Multi-Object Search using Object-Oriented POMDPs
This paper formulate a novel Multi-Object Search task as an OO-POMDP for mobile robotics domains in which the agent must find the locations of multiple objects, and develops a new algorithm for efficiently solving O-PomDPs: Object-Oriented Partially Observable Monte-Carlo Planning (OOPOMCP).
Online Planning for Target Object Search in Clutter under Partial Observability
This paper proposes Parameterized Action Partially Observable Monte-Carlo Planning (PA-POMCP), an algorithm that evaluates manipulation actions by taking into account the effect of the robot’s current belief on the success of the action execution.
Act to See and See to Act: POMDP planning for objects search in clutter
This work model the problem of objects search in clutter as a Partially Observable Markov Decision Process (POMDP), formulating it as a problem of optimal decision making under uncertainty, and is able to adapt online POMDP planners to handle objects search problems with large state space and action space.
Active Visual Object Search in Unknown Environments Using Uncertain Semantics
It is argued that by making use of uncertain semantics of the environment, a robot tasked with finding an object can devise efficient search strategies that can locate everyday objects at the scale of an entire building floor, which is previously unknown to the robot.
Manipulation-based active search for occluded objects
A novel generative model for representing container contents by using object co-occurrence information and spatial constraints is presented, which is used to guide an agent to explore containers where the target is likely, potentially needing to move occluding objects to enable further perception.
Efficient Object Search With Belief Road Map Using Mobile Robot
This letter introduces a novel scheme for the belief propagation when the robot is searching for an object as a Partially Observable Markov Decision Process (POMDP) and proposes a novel graph structure called belief road map (BRM) that is built along with the searching process.
Mechanical Search: Multi-Step Retrieval of a Target Object Occluded by Clutter
A version where distractor objects are heaped over the target object in a bin and success can be achieved in this long-horizon task with algorithmic policies in over 95% of instances and that the number of actions required scales approximately linearly with the size of the heap.
Nonmyopic View Planning for Active Object Classification and Pose Estimation
An active approach to object detection in which the point of view of a mobile depth camera is controlled and outperforms the widely used greedy viewpoint selection and provides a significant improvement over static object detection.
Monte-Carlo Planning in Large POMDPs
POMCP is the first general purpose planner to achieve high performance in such large and unfactored POMDPs as 10 x 10 battleship and partially observable PacMan, with approximately 1018 and 1056 states respectively.
Sensor planning in 3d object search: its formu-lation and complexity
This paper forms the task of sensor planning for object search from the theoretical point of view, and analyzes this task from the complexity level and proves that this task is NP-Complete.