European Commission
European Defence Fund
2023 – 2027
5 M€
Project website
AIDEDeX (Artificial Intelligence for Detection of Explosive Ordnance Extended) aims to provide a baseline for advanced and mature research for multi-robotic systems with State of the Art sensors and Artificial Intelligence Algorithms for the detection and classification of IEDs (Improvised Explosive Devices), EOs (Explosive Ordnance) and Landmines.
The concept of AIDEDeX relies on the strong foundations of the on-going project AIDED in which the consortium is carrying out the technology development, offering a solid base for an highly reliable platform in AIDEDeX. The main focus is on the automated detection of Improvised Explosive Devices (IEDs) using ArtificialIntelligence techniques on data collected from a large suite of sensors such as Electro-Magnetic Inference (EMI), Ground Penetrating Radar (GPR), X-Ray Backscatter Imaging (XRB), Raman Spectrometer, Infrared Cameras, Multispectral Cameras and Chemical. This set of sensors, or a combination of them, allows to determine the position and type of the IED and landmines with a maximum accuracy while minimising every risk.
A fleet of mixed Unmanned Ground Vehicles (UGV) and Aerial (UAV) will be able to operate in different scenarios, from open space fields to more risky closed environments and urban scenarios, of increasing interest for security in cities. The system will be able to plan the mission in complete autonomy, receiving just the perimeter of the area to explore. One of the UGV will also be equipped with a robotic arm for manipulation of the IED with different manipulation systems, such as pliers with force feedback to carefully grab the object, scissors to cut wires and air-compressed gun to clean eventual dust or cover on the ordnance. The core idea of this approach is to provide a fully autonomous intelligent system, with capabilities of merging and analysing data from heterogeneous sensors and decisional autonomy to operate on the environment, with performances that overcome the more consolidated approach of human operator and teleoperated robot. It would allow to overcome classical issues as environments with telecommunication denial and intrinsic difficulties in operating several robots of different dimensions and with different locomotion systems, while increasing at the same time the capacity of the operator to control the system, using hybrid interfaces between teleoperation and semi-autonomous algorithms.
The innovations resulting from the project include a complete, reliable robotics solution (mission planning, deploying, monitoring) compatible with a large range of applications.
Project Publications
2025
- E. Maroulis, D. Hawari, K. Hasselmann, E. {Le Flécher}, and G. {De cubber}, “Experimental Evaluation of Roadmap-Based Map Generation with Continuous-Time Conflict-Based Search for Multi-Agent Pathfinding," IEEE International Conference on Autonomous Robots and Agents, ICARA, p. 380–387, 2025.
[BibTeX] [Abstract] [Download PDF] [DOI]
This article presents an experimental evaluation of a Multi-Agent Pathfinding (MAPF) approach for large-scale robotic fleets operating in diverse outdoor environments. We generated three distinct types of roadmap graphs: Constrained Delaunay Triangulation (CDT), Voronoi Diagram (VD), and Grid-derived from an obstacle file, and assessed their quality using metrics obtained from graph theory. Then, the performance of the Continuous-time Conflict-Based Search (CCBS) algorithm was evaluated across three different environmental maps, considering practical performance metrics including makespan and failure rate. Subsequently, the roadmap generation methods were ranked based on CCBS performance in similar scenarios using the Friedman statistical test. The results indicate that CDT outperforms both VD and Grid maps, even though it does not exhibit the best graph metrics in many environments. CDT’s superior performance is attributed to its enhanced interconnectedness and the availability of multiple pathways, as evidenced by its balanced metrics and structural properties. We show that CDT is the most efficient and reliable roadmap generation technique for multiagent systems under our experimental conditions making it a preferred choice for robust pathfinding.
@article{34774d01cc3341398188fc8353028be2, title = "Experimental Evaluation of Roadmap-Based Map Generation with Continuous-Time Conflict-Based Search for Multi-Agent Pathfinding", abstract = "This article presents an experimental evaluation of a Multi-Agent Pathfinding (MAPF) approach for large-scale robotic fleets operating in diverse outdoor environments. We generated three distinct types of roadmap graphs: Constrained Delaunay Triangulation (CDT), Voronoi Diagram (VD), and Grid-derived from an obstacle file, and assessed their quality using metrics obtained from graph theory. Then, the performance of the Continuous-time Conflict-Based Search (CCBS) algorithm was evaluated across three different environmental maps, considering practical performance metrics including makespan and failure rate. Subsequently, the roadmap generation methods were ranked based on CCBS performance in similar scenarios using the Friedman statistical test. The results indicate that CDT outperforms both VD and Grid maps, even though it does not exhibit the best graph metrics in many environments. CDT's superior performance is attributed to its enhanced interconnectedness and the availability of multiple pathways, as evidenced by its balanced metrics and structural properties. We show that CDT is the most efficient and reliable roadmap generation technique for multiagent systems under our experimental conditions making it a preferred choice for robust pathfinding.", keywords = "Measurement , Automation , Reliability theory , Graph theory , Path planning , Robots , Multi-agent systems", author = "Emmanouil Maroulis and Danial Hawari and Ken Hasselmann and {Le Fl{\'e}cher}, Emile and {De cubber}, Geert", year = "2025", month = may, day = "5", doi = "10.1109/ICARA64554.2025.10977707", language = "English", pages = "380--387", journal = "IEEE International Conference on Autonomous Robots and Agents, ICARA", issn = "2767-7745", url = "https://ieeexplore.ieee.org/document/10977707", unit= {meca-ras}, project= {CUGS, ANIMUS, AIDEDEX, CONVOY} }
2024
- K. Hasselmann, M. Malizia, R. Caballero, F. Polisano, S. Govindaraj, J. Stigler, O. Ilchenko, M. Bajic, and G. De Cubber, “A multi-robot system for the detection of explosive devices," in “IEEE ICRA Workshop on Field Robotics", 2024.
[BibTeX] [Download PDF] [DOI]@inproceedings{Hasselmannetal2024ICRAWSFRO, doi = {10.48550/ARXIV.2404.14167}, url={https://arxiv.org/abs/2404.14167}, booktitle = {"IEEE ICRA Workshop on Field Robotics"}, author = {Hasselmann, Ken and Malizia, Mario and Caballero, Rafael and Polisano, Fabio and Govindaraj, Shashank and Stigler, Jakob and Ilchenko, Oleksii and Bajic, Milan and De Cubber, Geert}, title = {A multi-robot system for the detection of explosive devices}, year = {2024}, unit= {meca-ras}, project= {AIDED, AIDEDEX} } - M. Kegeleirs, D. G. Ramos, K. Hasselmann, L. Garattoni, G. Francesca, and M. Birattari, “Transferability in the automatic off-line design of robot swarms: from sim-to-real to embodiment and design-method transfer across different platforms," IEEE Robotics and Automation Letters, 2024.
[BibTeX] [Download PDF] [DOI]@article{kegeleirs2024transferability, title={Transferability in the automatic off-line design of robot swarms: from sim-to-real to embodiment and design-method transfer across different platforms}, author={Kegeleirs, Miquel and Ramos, David Garz{\'o}n and Hasselmann, Ken and Garattoni, Lorenzo and Francesca, Gianpiero and Birattari, Mauro}, journal={IEEE Robotics and Automation Letters}, year={2024}, doi={https://doi.org/10.1109/LRA.2024.3360013}, url={https://ieeexplore.ieee.org/document/10416330}, publisher={IEEE}, unit= {meca-ras}, project= {AIDEDEX} }