Fast or Slow: An Autonomous Speed Control Approach for UAV-assisted IoT Data Collection Networks

TL;DR

This paper presents an autonomous speed control method for UAVs in IoT data collection, using advanced deep reinforcement learning techniques to optimize energy efficiency and system performance.

Contribution

It introduces a novel deep dueling double Q-learning approach for UAV speed control, considering energy status and location, improving performance over traditional methods.

Findings

Achieves up to 40% performance improvement over conventional methods.

Highlights the impact of UAV energy and charging time on system efficiency.

Demonstrates the effectiveness of deep reinforcement learning in UAV speed management.

Abstract

Unmanned Aerial Vehicles (UAVs) have been emerging as an effective solution for IoT data collection networks thanks to their outstanding flexibility, mobility, and low operation costs. However, due to the limited energy and uncertainty from the data collection process, speed control is one of the most important factors to optimize the energy usage efficiency and performance for UAV collectors. This work aims to develop a novel autonomous speed control approach to address this issue. To that end, we first formulate the dynamic speed control task of a UAV as a Markov decision process taking into account its energy status and location. In this way, the Q-learning algorithm can be adopted to obtain the optimal speed control policy for the UAV. To further improve the system performance, we develop an highly-effective deep dueling double Q-learning algorithm utilizing outstanding features of the deep neural networks as well as advanced dueling architecture to quickly stabilize the learning process and obtain the optimal policy. Through simulation results, we show that our proposed solution can achieve up to 40% greater performance compared with other conventional methods. Importantly, the simulation results also reveal significant impacts of UAV's energy and charging time on the system performance.