Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HADRON: Human-friendly Control and Artificial Intelligence for Military Drone Operations

Ana M. Casado Faulí, Mario Malizia, Ken Hasselmann, Emile Le Flécher, Geert De Cubber, Ben Lauwens

TL;DR

HADRON tackles the challenge of enabling untrained personnel to operate military drone swarms while managing cognitive load and data interpretation. It proposes a three-tier autonomy framework and a modular interface suite (tablet, voice, gesture, AR) to support novice operators, expert supervision of fleets, and AI-assisted real-time processing. Key contributions include a system architecture that integrates human-friendly controls, multi-agent coordination, and semi-automated perception and defect-detection pipelines. The approach aims to enable scalable, safer, and faster deployment of UAS in complex environments with reduced need for extensive training.

Abstract

As drones are getting more and more entangled in our society, more untrained users require the capability to operate them. This scenario is to be achieved through the development of artificial intelligence capabilities assisting the human operator in controlling the Unmanned Aerial System (UAS) and processing the sensor data, thereby alleviating the need for extensive operator training. This paper presents the HADRON project that seeks to develop and test multiple novel technologies to enable human-friendly control of drone swarms. This project is divided into three main parts. The first part consists of the integration of different technologies for the intuitive control of drones, focusing on novice or inexperienced pilots and operators. The second part focuses on the development of a multi-drone system that will be controlled from a command and control station, in which an expert pilot can supervise the operations of the multiple drones. The third part of the project will focus on reducing the cognitive load on human operators, whether they are novice or expert pilots. For this, we will develop AI tools that will assist drone operators with semi-automated real-time data processing.

HADRON: Human-friendly Control and Artificial Intelligence for Military Drone Operations

TL;DR

HADRON tackles the challenge of enabling untrained personnel to operate military drone swarms while managing cognitive load and data interpretation. It proposes a three-tier autonomy framework and a modular interface suite (tablet, voice, gesture, AR) to support novice operators, expert supervision of fleets, and AI-assisted real-time processing. Key contributions include a system architecture that integrates human-friendly controls, multi-agent coordination, and semi-automated perception and defect-detection pipelines. The approach aims to enable scalable, safer, and faster deployment of UAS in complex environments with reduced need for extensive training.

Abstract

As drones are getting more and more entangled in our society, more untrained users require the capability to operate them. This scenario is to be achieved through the development of artificial intelligence capabilities assisting the human operator in controlling the Unmanned Aerial System (UAS) and processing the sensor data, thereby alleviating the need for extensive operator training. This paper presents the HADRON project that seeks to develop and test multiple novel technologies to enable human-friendly control of drone swarms. This project is divided into three main parts. The first part consists of the integration of different technologies for the intuitive control of drones, focusing on novice or inexperienced pilots and operators. The second part focuses on the development of a multi-drone system that will be controlled from a command and control station, in which an expert pilot can supervise the operations of the multiple drones. The third part of the project will focus on reducing the cognitive load on human operators, whether they are novice or expert pilots. For this, we will develop AI tools that will assist drone operators with semi-automated real-time data processing.
Paper Structure (8 sections, 2 figures, 1 table)

This paper contains 8 sections, 2 figures, 1 table.

Table of Contents

  1. INTRODUCTION
  2. STATE OF THE ART
  3. SYSTEM OVERVIEW
  4. Human-Friendly Control of a Drone by Untrained Users
  5. Multi-agent Control of Drones by Expert Users
  6. Semi-automated data interpretation
  7. ADVANCED PILOT TRAINING TOOLS
  8. CONCLUSIONS

Figures (2)

  • Figure 1: Depiction of a dismounted soldier operating a drone, using voice commands. The drone detects a target and informs the operator, via audio, so the operator can make decisions on the spot.
  • Figure 2: Depiction of a command officer operating a UAS patrol system, using voice commands. After a patrol mission has been deployed, if the drones detect an intruder, they identify it and inform the operator about it.