An Efficient Deep Reinforcement Learning Model for Online 3D Bin Packing Combining Object Rearrangement and Stable Placement

This paper presents an efficient deep reinforcement learning (DRL) framework for online 3D bin packing (3D-BPP). The 3D-BPP is an NP-hard problem significant in logistics, warehousing, and transportation, involving the optimal arrangement of objects inside a bin. Traditional heuristic algorithms oft...

Full description

Saved in:
Bibliographic Details
Published inInternational Conference on Control, Automation and Systems (Online) pp. 964 - 969
Main Authors Zhou, Peiwen, Gao, Ziyan, Li, Chenghao, Chong, Nak Young
Format Conference Proceeding
LanguageEnglish
Published ICROS 29.10.2024
Subjects
3D Bin Packing
Deep reinforcement learning
Heuristic algorithms
Object Rearrangement
Physics
Placement Stability
Reliability
Stability analysis
Testing
Three-dimensional displays
Training
Transportation
Warehousing
Online AccessGet full text
ISSN2642-3901
DOI10.23919/ICCAS63016.2024.10773090

Cover

More Information
Summary:This paper presents an efficient deep reinforcement learning (DRL) framework for online 3D bin packing (3D-BPP). The 3D-BPP is an NP-hard problem significant in logistics, warehousing, and transportation, involving the optimal arrangement of objects inside a bin. Traditional heuristic algorithms often fail to address dynamic and physical constraints in real-time scenarios. We introduce a novel DRL framework that integrates a reliable physics heuristic algorithm and object rearrangement and stable placement. Our experiment show that the proposed framework achieves higher space utilization rates effectively minimizing the amount of wasted space with fewer training epochs.
ISSN:2642-3901
DOI:10.23919/ICCAS63016.2024.10773090