.png){kind=icon}
Return
Self Evolving Mobile Charging Scheduling
.png){kind=icon}
PRE
AI
DOI：10.1109/tmc.2026.3728442.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Wireless Sensor Networks (WSNs) have become popular due to their lightweight infrastructure and autonomous configuration, although the limited power supply poses a challenge to continuous operation. Wireless Rechargeable Sensor Networks (WRSNs), supported by Wireless Power Transfer (WPT), provide a feasible solution for continuous monitoring. Nevertheless, unpredictable topology changes, such as sensor activation, deactivation, and repositioning, directly impact charging efficiency and path planning. To address this, we focus on the mobile charging scheduling problem in such Dynamic Wireless Rechargeable Sensor Networks (DWRSNs). We formulate the problem as a charging utility Maximization problem IN DWRSNs (MIND) and propose a Hierarchical Self-evolving Charging Optimization (HSCO) algorithm based on Hierarchical Reinforcement Learning (HRL). HSCO leverages Transformer architectures with a Proximal Policy Optimization (PPO)-based upper-level policy for periodically detecting network dynamics and charging region division, and a REINFORCE-based lower-level policy for generating local paths with real-time strategy evolution in response to network topology changes. The simulation and experimental results demonstrate that HSCO achieves strong adaptability, high efficiency, and good scalability, improving charging utility by an average of 121.04%.
Keywords:
Wireless power transfer
wireless rechargeable sensor networks
mobile charging
reinforcement learning
Journal
IF:
9.2
Papers:
5.8K
Citations:
1.8W
Organization
View more organizations
Cited Papers
No cited papers available