Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Attention-based UAV Trajectory Optimization for Wireless Power Transfer-assisted IoT Systems

Li Dong, Feibo Jiang, Yubo Peng

TL;DR

The paper addresses energy-efficient UAV trajectory optimization in WPT-assisted IoT systems with limited UAV energy and IoTD storage. It introduces AUTO, a framework that combines a graph-transformer based ATOM for encoding/decoding IoTDs and a stable TENMA-based training regime for large-scale trajectory planning. The authors demonstrate that ATOM-TENMA achieves lower energy consumption and feasible, scalable trajectories across simulations and a field case, outperforming several baselines and neural-network-based designers. This work provides a practical, scalable approach to joint UAV deployment, association, and scheduling in WPT-IoT networks, with strong potential for real-world deployment.

Abstract

Unmanned Aerial Vehicles (UAVs) in Wireless Power Transfer (WPT)-assisted Internet of Things (IoT) systems face the following challenges: limited resources and suboptimal trajectory planning. Reinforcement learning-based trajectory planning schemes face issues of low search efficiency and learning instability when optimizing large-scale systems. To address these issues, we present an Attention-based UAV Trajectory Optimization (AUTO) framework based on the graph transformer, which consists of an Attention Trajectory Optimization Model (ATOM) and a Trajectory lEarNing Method based on Actor-critic (TENMA). In ATOM, a graph encoder is used to calculate the self-attention characteristics of all IoTDs, and a trajectory decoder is developed to optimize the number and trajectories of UAVs. TENMA then trains the ATOM using an improved Actor-Critic method, in which the real reward of the system is applied as the baseline to reduce variances in the critic network. This method is suitable for high-quality and large-scale multi-UAV trajectory planning. Finally, we develop numerous experiments, including a hardware experiment in the field case, to verify the feasibility and efficiency of the AUTO framework.

Attention-based UAV Trajectory Optimization for Wireless Power Transfer-assisted IoT Systems

TL;DR

The paper addresses energy-efficient UAV trajectory optimization in WPT-assisted IoT systems with limited UAV energy and IoTD storage. It introduces AUTO, a framework that combines a graph-transformer based ATOM for encoding/decoding IoTDs and a stable TENMA-based training regime for large-scale trajectory planning. The authors demonstrate that ATOM-TENMA achieves lower energy consumption and feasible, scalable trajectories across simulations and a field case, outperforming several baselines and neural-network-based designers. This work provides a practical, scalable approach to joint UAV deployment, association, and scheduling in WPT-IoT networks, with strong potential for real-world deployment.

Abstract

Unmanned Aerial Vehicles (UAVs) in Wireless Power Transfer (WPT)-assisted Internet of Things (IoT) systems face the following challenges: limited resources and suboptimal trajectory planning. Reinforcement learning-based trajectory planning schemes face issues of low search efficiency and learning instability when optimizing large-scale systems. To address these issues, we present an Attention-based UAV Trajectory Optimization (AUTO) framework based on the graph transformer, which consists of an Attention Trajectory Optimization Model (ATOM) and a Trajectory lEarNing Method based on Actor-critic (TENMA). In ATOM, a graph encoder is used to calculate the self-attention characteristics of all IoTDs, and a trajectory decoder is developed to optimize the number and trajectories of UAVs. TENMA then trains the ATOM using an improved Actor-Critic method, in which the real reward of the system is applied as the baseline to reduce variances in the critic network. This method is suitable for high-quality and large-scale multi-UAV trajectory planning. Finally, we develop numerous experiments, including a hardware experiment in the field case, to verify the feasibility and efficiency of the AUTO framework.

Paper Structure

This paper contains 28 sections, 32 equations, 5 figures, 6 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. Trajectory Model
  4. Data Collection Model
  5. Energy Consumption Model
  6. Problem Formulation
  7. Problem Decomposition
  8. Attention-based UAV Trajectory Optimization Framework
  9. AUTO Framework Overview
  10. Graph Encoder
  11. Graph Embedding Calculation
  12. Self Attention Calculation
  13. Graph Pooling Calculation
  14. Trajectory Decoder
  15. Graph State Definition
  16. ...and 13 more sections

Figures (5)

  • Figure 1: The WPT-assisted IoT system.
  • Figure 2: Structure of the graph encoder.
  • Figure 3: The trajectories generated by the trajectory decoder.
  • Figure 4: Energy cost with different battery capacities of UAVs.
  • Figure 5: The generated trajectories of UAVs in the field case.