Explainable, Safe, Contact-Aware Planning and Control for Heavy Machinery Manipulation and Navigation

Informazioni relative al progetto

XSCAVE

ID dell’accordo di sovvenzione: 101189836

DOI

10.3030/101189836

Data della firma CE 4 Novembre 2024

Data di avvio 1 Dicembre 2024

Data di completamento 30 Novembre 2028

Finanziato da

• Digital, Industry and Space

Costo totale

€ 7 975 530,00

Contributo UE

€ 7 975 530,00

Coordinato da

TARTU ULIKOOL
Estonia

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Documents, reports (9)

Data and ethics management plan

The deliverable pertains to developing a detailed plan to consider the ethical aspects of the project, especially with regards to the use of AI. It will also cover a systematic plan to manage different types of data generated within the project

Requirement, use case, detailed plan document 

The report will provide a survey of the key technological challenges, gaps and requirements of different use-cases of the project. The report will serve as the foundation for fine-tuning the algorithmic research of the project in order to better alignn with the KPIs of different use-cases.

Dissemination and Communication Plan

The report will summarize the strategy that will be adopted in the project for dissemination and communication of the results and achievements.

Trustworthy AI guidance of the XSCAVE

The delivered report will compile a guideline for the consortium partners to ensure that the principles of trustworthy AI and the Ethical, Legal, Socio-Economic and Environmental, i.e., ELSE factors are considered throughout the project execution

Tools for generating causality guided explanations from deep neural model-based planners

The report will provide a documentation of the tool developed for generating causal explaination of neural planners. An open-source code-base will also be released along with the report

Specification of interfaces to ensure compatibility; Prototype version of pipeline with preliminary components 

This report will cover a survey of all the different simulation and experimental interfaces that will be necessary for the execution of the core research of the project. It will also cover a generic preliminary pipeline for perception to action mapping that will also serve as the foundation for developing more specific pipelines for each use-cases.

Prototype version of learned neural planners and controllers with risk-aware optimization priors

The report will summarize the algorithms and code-base for the first version of neural planners that perform risk-aware decision making to counter the uncertainty of sim2real

Prototype version of perception module with mock-up/preliminary components

The report will summarize the algorithm/software and code-base of the first version of the perception pipeline developed from off-the-shelf components and that based on prior research of the partners

First Deployable Version of the entire pipeline to be deployed by the end users

The report will summarize the first version of the end-to-end perception to control pipeline for different use-cases considered in the project. The report will also be accompanied with open-source software.

Other (1)

Visuals and branding materials, Website

This deliverable pertains to creating a website and other branding materials for publicizing the project and its activities in different projects.

Pubblicazioni

Peer reviewed articles (3)

Physics-informed data-driven modeling of rock motion dynamics in excavation using a high-fidelity simulator

Autori: Mohammad Heravi, Amirmasoud Molaei, Reza Ghabcheloo
Pubblicato in: Simulation Modelling Practice and Theory, Numero 145, 2026, ISSN 1569-190X
Editore: Elsevier BV
DOI: 10.1016/J.SIMPAT.2025.103208

FORWARD: Dataset of a forwarder operating in rough terrain

Autori: Mikael Lundbäck, Erik Wallin, Carola Häggström, Mattias Nyström, Andreas Grönlund, Mats Richardson, Petrus Jönsson, William Arnvik, Lucas Hedström, Arvid Fälldin, Martin Servin
Pubblicato in: Data in Brief, Numero 66, 2026, ISSN 2352-3409
Editore: Elsevier BV
DOI: 10.1016/J.DIB.2026.112725

Trajectory Optimization Under Stochastic Dynamics Leveraging Maximum Mean Discrepancy

Autori: Basant Sharma, Arun Kumar Singh
Pubblicato in: IEEE Robotics and Automation Letters, Numero 10, 2025, ISSN 2377-3766
Editore: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2025.3565335

Thesis and dissertations (2)

Reality to Simulation: A Scene Understanding Approach to 3D Log Pile Scene Reconstruction

Autori: Lundberg, Philiph
Pubblicato in: 2025
Editore: UMU

Simulation of Heavy Vehicle Traversal on Deformable Terrain

Autori: Edvin, Hortlund
Editore: UMU

Conference proceedings (5)

Manipulate-To-Navigate: Reinforcement Learning with Visual Affordances and Manipulability Priors

Autori: Yuying Zhang, Joni Pajarinen
Pubblicato in: 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025
Editore: IEEE
DOI: 10.1109/IROS60139.2025.11247222

Manual, Semi or Fully Autonomous Flipper Control? A Framework for Fair Comparison

Autori: Valentýn Číhala, Martin Pecka, Tomáš Svoboda, Karel Zimmermann
Pubblicato in: 2025 IEEE International Conference on Robotics and Automation (ICRA), 2025
Editore: IEEE
DOI: 10.1109/ICRA55743.2025.11128663

Minimally Conservative Controlled-Invariant Set Synthesis Using Control Barrier Certificates

Autori: Naeim Ebrahimi Toulkani, Reza Ghabcheloo
Pubblicato in: 2025 IEEE 64th Conference on Decision and Control (CDC), 2026
Editore: IEEE
DOI: 10.1109/CDC57313.2025.11312766

Extracting Visual Plans from Unlabeled Videos via Symbolic Guidance

Autori: Wenyan Yang, Ahmet Tikna, Yi Zhao, Yuying Zhang, Luigi Palopoli, Marco Roveri, Joni Pajarinen
Pubblicato in: 2025
Editore: PMLR

Ground Vehicle Detection Metrics And Terramechanical Characterization Derived From A Distributed Acoustic Sensing System Deployed On An Arctic Ice Road

Autori: Aaron Meyer, Adrian Doran, Abhishek Bhagat, Meghan Quinn, Keith Wilson, Matthew Kimball, Jennifer Picucci, Wei Huang, Russ Alger
Pubblicato in: Proceedings of the 55th Conference of the ISTVS, 2026
Editore: International Society for Terrain-Vehicle Systems
DOI: 10.56884/ISF3IT7O