Multi-robot path planning based on a deep reinforcement learning DQN algorithm

The unmanned warehouse dispatching system of the ‘goods to people’ model uses a structure mainly based on a handling robot, which saves considerable manpower and improves the efficiency of the warehouse picking operation. However, the optimal performance of the scheduling system algorithm has high r...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:CAAI Transactions on Intelligence Technology Jg. 5; H. 3; S. 177 - 183
Hauptverfasser: Yang, Yang, Juntao, Li, Lingling, Peng
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Beijing The Institution of Engineering and Technology 01.09.2020
John Wiley & Sons, Inc
Wiley
Schlagworte:
algorithmic process
Algorithms
Automation
classic deep reinforcement learning algorithm
Decision making
Deep learning
deep Q‐network algorithm
handling robot
improved DQN algorithm converges
learning (artificial intelligence)
Machine learning
mobile robots
multi-robot systems
multirobot path planning
Neural networks
Path planning
path-planning problems
Q‐learning algorithm
Research Article
robot path-planning problem
Robots
scheduling system algorithm
unmanned warehouse dispatching system
volume-based technology
warehouse picking operation
Warehouses
warehouse picking operation
scheduling system algorithm
multirobot path planning
multi-robot systems
unmanned warehouse dispatching system
path-planning problems
algorithmic process
mobile robots
path planning
robot path-planning problem
volume-based technology
improved DQN algorithm converges
learning (artificial intelligence)
Q-learning algorithm
classic deep reinforcement learning algorithm
deep Q-network algorithm
handling robot
ISSN:2468-2322, 2468-6557, 2468-2322
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The unmanned warehouse dispatching system of the ‘goods to people’ model uses a structure mainly based on a handling robot, which saves considerable manpower and improves the efficiency of the warehouse picking operation. However, the optimal performance of the scheduling system algorithm has high requirements. This study uses a deep Q-network (DQN) algorithm in a deep reinforcement learning algorithm, which combines the Q-learning algorithm, an empirical playback mechanism, and the volume-based technology of productive neural networks to generate target Q-values to solve the problem of multi-robot path planning. The aim of the Q-learning algorithm in deep reinforcement learning is to address two shortcomings of the robot path-planning problem: slow convergence and excessive randomness. Preceding the start of the algorithmic process, prior knowledge and prior rules are used to improve the DQN algorithm. Simulation results show that the improved DQN algorithm converges faster than the classic deep reinforcement learning algorithm and can more quickly learn the solutions to path-planning problems. This improves the efficiency of multi-robot path planning.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2468-2322
2468-6557
2468-2322
DOI:10.1049/trit.2020.0024