Tunnel construction is a dynamic excavation process that requires a variety of construction equipment and workers to cooperate according to standard procedures. To enhance construction efficiency and reduce labor costs, intelligent construction equipments have been extensively employed in tunnel construction sites. However, the effectiveness of collaborative operation efficiency in construction equipment is hindered to some extent due to inadequate information sharing and inefficient awareness of the construction situation. To deal with the problem, we propose a cluster-based communication network architecture and define a novel metric named construction procedure coherence degree (CPCD) based on age of information (AoI) to quantify the significance of construction equipment collaborative operation. We formulate radio resource allocation as a stochastic optimization problem, which can be modeled as a decentralized partially observable Markov decision process (Dec-POMDP). Furthermore, a multi-agent reinforcement learning algorithm named QMIX is used to optimize the spectrum-sharing strategies among multiple construction equipment, enabling efficient and reliable communication. Extensive experiment results show that the QMIX outperforms well than independent Dueling DQN (IDDQN).
AoI-Driven Cluster-Based Spectrum Sharing in Internet of Construction Equipment with Multi-Agent Reinforcement Learning
24.09.2023
410797 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
DOAJ | 2025
|Distributed Cooperative Spectrum Sharing in UAV Networks Using Multi-Agent Reinforcement Learning
ArXiv | 2018
|Multi-agent reinforcement learning traffic signal control method based on neighbor knowledge sharing
Europäisches Patentamt | 2023
|