3D localization of machines and tools based on deep reinforcement learning

Ruikai Zhu; Ziqiang Lu; Kai Sun; Liangan Yao; Aisheng Yang; Xianhong Jiao; Zhou Shen

doi:10.1117/12.2620293

29 November 2021 3D localization of machines and tools based on deep reinforcement learning

Ruikai Zhu, Ziqiang Lu, Kai Sun, Liangan Yao, Aisheng Yang, Xianhong Jiao, Zhou Shen

Proceedings Volume 12080, 4th International Symposium on Power Electronics and Control Engineering (ISPECE 2021); 120800E (2021) https://doi.org/10.1117/12.2620293
Event: 4th International Symposium on Power Electronics and Control Engineering (ISPECE 2021), 2021, Nanchang, China

Abstract

In view of the importance of obtaining the position information of machines and tools in the collection and safety control of incoming and outgoing information of power equipment, a wireless three-dimensional positioning method based on deep reinforcement learning DRL is proposed. Modeling is based on the signal non-line-of-sight(NLOS) propagation model, which will be based on measurement. The process of target location features such as angle (AOD), angle of arrival (AOA) and time of arrival (TOA) is modeled as a Markov decision process and processed within the DRL framework by using the location features of the above targets as input Finally, the repeated training deep Q-network(DQN) algorithm is used for predictive positioning. The simulation results show that the advantages of using DRL in wireless positioning are verified.

Citation Download Citation

Ruikai Zhu, Ziqiang Lu, Kai Sun, Liangan Yao, Aisheng Yang, Xianhong Jiao, and Zhou Shen "3D localization of machines and tools based on deep reinforcement learning", Proc. SPIE 12080, 4th International Symposium on Power Electronics and Control Engineering (ISPECE 2021), 120800E (29 November 2021); https://doi.org/10.1117/12.2620293

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available