Spatial Perception and Embodied Autonomy Research

Informations projet

SPEAR

N° de convention de subvention: 101119774

DOI

10.3030/101119774

Date de signature de la CE 27 Juin 2023

Date de début 1 Novembre 2023

Date de fin 31 Octobre 2027

Financé au titre de

• Digital, Industry and Space

Coût total

€ 6 746 086,25

Contribution de l’UE

€ 6 746 086,25

Coordonné par

NORGES TEKNISK-NATURVITENSKAPELIGE UNIVERSITET NTNU
Norway

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Documents, reports (10)

Evolutionary flying robot embodiment generation

Report on evolutionary Human-&-AI embodiment co-generation algorithms. The report shall contain also software examples.

Ultra-low latency multi-modal SLAM

Report on low-latency event-based active ranging, ULL (multi-modal) SLAM. The report shall contain also software examples.

Navigation policy learning and constraint handling

Report on baseline navigation policy learning for automatically generated flying embodiments with constraint handling. The report shall contain also software examples.

Final Simulation Toolbox

Report for the simulation toolbox. The report shall contain also software examples.

Collision detection and propulsion technologies for soft drones

Report on collision detection techniques and approaches on embedding propulsion for the generated bodies.

Dissemination, exploitation and communication plan

Dissemination, exploitation and communication plan report (iteratively updated).

Soft aerial robot design

Report on soft aerial robots design and embodiment sensing. The report shall contain also design examples.

Dissemination, exploitation and communication plan - Update 1

Dissemination, exploitation and communication plan report (iteratively updated).

Ultra-low latency multi-modal SLAM - Update

Updated report on low-latency event-based active ranging, ULL (multi-modal) SLAM. The report shall contain also software examples.

Mid-term Simulation Toolbox

Report for the first version of Simulation Toolbox. The report shall contain also software examples.

Other (2)

Software and Dataset releases

SPEAR Software and Dataset release involving the simulation toolbox (M18 - this deliverable), the soft materials design tool (M24 - Update 1), the Evolutionary Algorithms-based embodiment generation (M34 - Update 2), and the Navigation Policy Learning software (M44 - Update 3).

Software and Dataset releases - Update 1

SPEAR Software and Dataset release involving the simulation toolbox (M18), the soft materials design tool (M24 - this deliverable), the Evolutionary Algorithms-based embodiment generation (M34 - Update 2), and the Navigation Policy Learning software (M44 - Update 3).

Data Management Plan (2)

Data Management Plan

The DMP contains detailed procedures for all data types producing during SPEAR. The DMP will be a living document (see also planned updates).

Data Management Plan - Update 1

SPEAR Data Management Plan - Update 1

Demonstrators, pilots, prototypes (1)

Field testing and demonstration

Report on field test-based demonstration of the SPEAR prototypes. The report shall also contain data from the field tests.

Publications

Conference proceedings (15)

Reinforcement Learning for Active Perception in Autonomous Navigation

Auteurs: Grzegorz Malczyk, Mihir Kulkarni, Kostas Alexis
Publié dans: Accepted at IEEE ICRA 2026, 2026
Éditeur: IEEE
DOI: 10.48550/ARXIV.2602.01266

Curriculum-based Sample Efficient Reinforcement Learning for Robust Stabilization of a Quadrotor

Auteurs: Fausto Mauricio Lagos Suarez; Akshit Saradagi; Vidya Sumathy; Shruti Kotpalliwar; George Nikolakopoulos
Publié dans: CoRR, 2025
Éditeur: IEEE
DOI: 10.48550/ARXIV.2501.18490

Tightly-Coupled Radar-Visual-Inertial Odometry

Auteurs: Morten Nissov, Mohit Singh, Kostas Alexis
Publié dans: Accepted to IEEE ECC 2026, 2026
Éditeur: IEEE
DOI: 10.48550/ARXIV.2603.23052

A Graph-Based Reinforcement Learning Approach with Frontier Potential Based Reward for Safe Cluttered Environment Exploration

Auteurs: Gabriele Calzolari, Vidya Sumathy, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: 24th European Control Conference, 2026
Éditeur: IEEE

Belief Scene Graphs: Expanding Partial Scenes with Objects through Computation of Expectation

Auteurs: Mario A.V. Saucedo, Akash Patel, Akshit Saradagi, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024
Éditeur: IEEE
DOI: 10.48550/arXiv.2402.03840

Platform-Agnostic Reinforcement Learning Framework for Safe Exploration of Cluttered Environments with Graph Attention

Auteurs: Gabriele Calzolari, Vidya Sumathy, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: The 2026 IEEE International Conference on Robotics & Automation (ICRA)
Éditeur: IEEE

Decentralized Multi-Agent Reinforcement Learning Exploration with Inter-Agent Communication-Based Action Space

Auteurs: Gabriele Calzolari, Vidya Sumathy, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: 2024 24th International Conference on Control, Automation and Systems (ICCAS), 2024, ISSN 2642-3901
Éditeur: IEEE
DOI: 10.23919/ICCAS63016.2024.10773383

D-MARL: A Dynamic Communication-Based Action Space Enhancement for Multi Agent Reinforcement Learning Exploration of Large Scale Unknown Environments

Auteurs: Gabriele Calzolari, Vidya Sumathy, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2024

A Heuristic Approach for Performance Tuning in RL-based Quadrotor Control via Reward Design and Termination Conditions

Auteurs: Fausto Mauricio Lagos Suarez, Akshit Saradagi, Vidya Sumathy, George Nikolakopoulos
Publié dans: 34th Mediterranean Conference on Control and Automation (MED26), 2026
Éditeur: IEEE

Learning UAV Navigation with Collision-Aware Predicted-Motion Embedding

Auteurs: Mohammad Javad Zallaghi, Bryan Convens, Kelly Merckaert, Adrian Munteanu, Bram Vanderborght
Publié dans: 2026
Éditeur: Benelux Meeting on Systems and Control 2026

Fuzzy Logic Theory-based Adaptive Reward Shaping for Robust Reinforcement Learning (FARS)

Auteurs: Hürkan Şahin, Van Huyen Dang, Erdi Sayar, Alper Yegenoglu, Erdal Kayacan
Publié dans: IEEE WCCI 2026, 2026
Éditeur: IEEE
DOI: 10.48550/ARXIV.2604.15772

Thermal Image Refinement with Depth Estimation using Recurrent Networks for Monocular ORB-SLAM3

Auteurs: Hürkan Şahin, Huy Xuan Pham, Van Huyen Dang, Alper Yegenoglu, Erdal Kayacan
Publié dans: IEEE International Conference on Robotics & Automation, 2026
Éditeur: IEEE

Vine Robots that Evert through Bending

Auteurs: Rui Wu, Stefano Mintchev
Publié dans: 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024
Éditeur: IEEE
DOI: 10.3929/ethz-b-000705656

UniPilot: Enabling GPS-Denied Autonomy Across Embodiments

Auteurs: Mihir Kulkarni, Mihir Dharmadhikari, Nikhil Khedekar, Morten Nissov, Mohit Singh, Philipp Weiss, Kostas Alexis
Publié dans: 2025 IEEE International Conference on Advanced Robotics (ICAR), 2026
Éditeur: IEEE
DOI: 10.1109/ICAR65334.2025.11338632

Leveraging Computation of Expectation Models for Commonsense Affordance Estimation on 3D Scene Graphs

Auteurs: Mario A.V. Saucedo, Nikolaos Stathoulopoulos, Akash Patel, Christoforos Kanellakis, George Nikolakopoulos
Publié dans: 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024
Éditeur: IEEE
DOI: 10.48550/arXiv.2409.05392

Peer reviewed articles (12)

Efficient Knowledge Transfer for Jump-Starting Control Policy Learning of Multirotors Through Physics-Aware Neural Architectures

Auteurs: Welf Rehberg, Mihir Kulkarni, Philipp Weiss, Kostas Alexis
Publié dans: IEEE Robotics and Automation Letters, Numéro 11, 2026, ISSN 2377-3766
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2026.3668585

VDS-Nav: Volumetric Depth-Based Safe Navigation for Aerial Robots–Bridging the Sim-to-Real Gap

Auteurs: Van Huyen Dang; Adrian Redder; Huy Xuan Pham; Andriy Sarabakha; Erdal Kayacan
Publié dans: IEEE Robotics and Automation Letters, 2025, ISSN 2377-3766
Éditeur: IEEE
DOI: 10.1109/LRA.2025.3606806

Morphy: A Compliant and Morphologically Aware Flying Robot

Auteurs: Paolo De Petris; Morten Nissov; Kostas Alexis
Publié dans: Advanced Intelligent Systems, 2024, ISSN 2640-4567
Éditeur: Wiley
DOI: 10.1002/aisy.202400493

Aerial Gym Simulator: A Framework for Highly Parallelized Simulation of Aerial Robots

Auteurs: Mihir Kulkarni; Welf Rehberg; Kostas Alexis
Publié dans: IEEE Robotics and Automation Letters, 2025, ISSN 2377-3766
Éditeur: IEEE
DOI: 10.48550/ARXIV.2503.01471

MAVRL: Learn to Fly in Cluttered Environments With Varying Speed

Auteurs: Hang Yu, ChristopheDe Wagter, Guido C. H. E de Croon
Publié dans: IEEE Robotics and Automation Letters, Numéro 10, 2025, ISSN 2377-3766
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2024.3522778

Collaborative Exploration With a Marsupial Ground-Aerial Robot Team Through Task-Driven Map Compression

Auteurs: Angelos Zacharia, Mihir Dharmadhikari, Kostas Alexis
Publié dans: IEEE Robotics and Automation Letters, Numéro 10, 2025, ISSN 2377-3766
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2025.3609040

Neural NMPC through signed distance field encoding for collision avoidance

Auteurs: Martin Jacquet, Marvin Harms, Kostas Alexis
Publié dans: The International Journal of Robotics Research, 2026, ISSN 0278-3649
Éditeur: SAGE Publications
DOI: 10.1177/02783649251401223

Depth Transfer: Learning to See Like a Simulator for Real-World Drone Navigation

Auteurs: Hang Yu, Christophe De Wagter, Guido C. H. E de Croon
Publié dans: IEEE Robotics and Automation Letters, Numéro 10, 2025, ISSN 2377-3766
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2025.3617729

Encoding mechanical intelligence using ultraprogrammable joints

Auteurs: Rui Wu, Luca Girardi, Stefano Mintchev
Publié dans: Science Advances, Numéro 11, 2025, ISSN 2375-2548
Éditeur: American Association for the Advancement of Science (AAAS)
DOI: 10.1126/SCIADV.ADV2052

Semantically-Driven Deep Reinforcement Learning for Inspection Path Planning

Auteurs: Grzegorz Malczyk; Mihir Kulkarni; Kostas Alexis
Publié dans: IEEE Robotics and Automation Letters, 2025, ISSN 2377-3766
Éditeur: IEEE
DOI: 10.48550/ARXIV.2505.14443

On the Characterization and Limits of 4D Radar for Aided Inertial Navigation

Auteurs: Morten Nissov, Kostas Alexis
Publié dans: IEEE Transactions on Field Robotics, 2026, ISSN 2997-1101
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/TFR.2026.3690584

A Terminal State Feasibility Governor for Real-Time Nonlinear Model Predictive Control Over Arbitrary Horizons

Auteurs: Bryan Convens; Dominic Liao-McPherson; Kelly Merckaert; Bram Vanderborght; Marco M. Nicotra
Publié dans: IEEE Transactions on Control Systems Technology, 2024, ISSN 1558-0865
Éditeur: IEEE
DOI: 10.1109/tcst.2024.3365996

Book chapters (1)

Unconventional Hexacopters via Evolution and Learning: Performance Gains and New Insights

Auteurs: Jed R. Muff, Keiichi Ito, Elijah H. W. Ang, Karine Miras, A. E. Eiben
Publié dans: Lecture Notes in Computer Science, Applications of Evolutionary Computation, 2026
Éditeur: Springer Nature Switzerland
DOI: 10.1007/978-3-032-23604-3_5

Other (1)

MAC: Morphology-Aware Control Design with Deep Reinforcement Learning for Navigation of Squeezable Unmanned Aerial Vehicles

Auteurs: Van Huyen Dang, Adrian Redder, Erdal Kayacan
Publié dans: Workshop on Morphology-Aware Policy and Design Learning (MAPoDeL) at CoRL 2024., 2024
Éditeur: OpenReview