Centralized DQN-Based Optimal Path Control of Multiple Robots 


Vol. 48,  No. 4, pp. 449-456, Apr.  2023
10.7840/kics.2023.48.4.449


PDF
  Abstract

With the recent advent of the 4th Industrial Revolution era, research on the use of autonomous and intelligent robots in a smart factory has been actively investigated. However, existing automatic guided vehicles (AGVs) still have many problems, such as the cost of infrastructure installation and difficulty in route modification, because of their significantly limited mobility. Therefore, we herein propose an optimal logistics robot control scheme based on a centralized deep Q-network (DQN) so that multiple robots can minimize collisions and get to their desired destinations in the shortest path. Specifically, a simulation environment similar to a simplified factory logistics warehouse is considered, and offline reinforcement learning of logistics robots is conducted in this environment. In addition, we demonstrate the performance excellency of the proposed DQN model so that the robots can arrive at their destinations by the optimal routes. Through the test based on hardware implementation, we show that multiple logistics robots can arrive at their desired destinations along optimal paths while minimizing collisions.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

JiminJeon, HyebinCho, SeungminLee, HowonLee, "Centralized DQN-Based Optimal Path Control of Multiple Robots," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 4, pp. 449-456, 2023. DOI: 10.7840/kics.2023.48.4.449.

[ACM Style]

JiminJeon, HyebinCho, SeungminLee, and HowonLee. 2023. Centralized DQN-Based Optimal Path Control of Multiple Robots. The Journal of Korean Institute of Communications and Information Sciences, 48, 4, (2023), 449-456. DOI: 10.7840/kics.2023.48.4.449.

[KICS Style]

JiminJeon, HyebinCho, SeungminLee, HowonLee, "Centralized DQN-Based Optimal Path Control of Multiple Robots," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 4, pp. 449-456, 4. 2023. (https://doi.org/10.7840/kics.2023.48.4.449)
Vol. 48, No. 4 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880