Semantic-Aware UAV Command and Control for Efficient IoT Data Collection
2026-04-09 • Robotics
Robotics
AI summaryⓘ
The authors present a system where drones collect images from IoT devices more efficiently. Each device compresses its image into a small, smart representation that can still be understood even if only parts are received. A base station sends commands to control the drone's path, aiming to stay close enough to devices for good image quality within a limited time. They use a type of machine learning called Double Deep Q-Learning to decide the best drone movements. Their tests show this method works better than simple strategies for covering devices and getting clearer images.
Unmanned Aerial Vehicles (UAVs)Internet of Things (IoT)Semantic CommunicationDeep Joint Source-Channel Coding (DeepJSCC)Markov Decision Process (MDP)Double Deep Q-Learning (DDQN)Base Station (BS)Image ReconstructionCommand-and-Control (C&C)Trajectory Optimization
Authors
Assane Sankara, Daniel Bonilla Licea, Hajar El Hammouti
Abstract
Unmanned Aerial Vehicles (UAVs) have emerged as a key enabler technology for data collection from Internet of Things (IoT) devices. However, effective data collection is challenged by resource constraints and the need for real-time decision-making. In this work, we propose a novel framework that integrates semantic communication with UAV command-and-control (C&C) to enable efficient image data collection from IoT devices. Each device uses Deep Joint Source-Channel Coding (DeepJSCC) to generate a compact semantic latent representation of its image to enable image reconstruction even under partial transmission. A base station (BS) controls the UAV's trajectory by transmitting acceleration commands. The objective is to maximize the average quality of reconstructed images by maintaining proximity to each device for a sufficient duration within a fixed time horizon. To address the challenging trade-off and account for delayed C&C signals, we model the problem as a Markov Decision Process and propose a Double Deep Q-Learning (DDQN)-based adaptive flight policy. Simulation results show that our approach outperforms baseline methods such as greedy and traveling salesman algorithms, in both device coverage and semantic reconstruction quality.