CORDIS - Risultati della ricerca dell’UE
CORDIS

Optimising Robot Performance while Dreaming

Risultati finali

Mobile robot manipulation use case analysis

This deliverable contains the specification and documentation of the mobile robot manipulation scenarios as a result of the use case analysis

Full version of the navigation portfolio and evaluation

The full version of the navigation portfolio is made available and evaluated near the end of the project It will also be made public for maximising the impact of the developments achieved in the project

Parameter optimization for task planners (OLP and reactive grasp planning)

One of the planning work package features is the parameter optimisation for task planners OLP and reactive grasp planning the further development and the connection of these features is reported in this deliverable

Parameter optimization for motion planners

The motion planners in the portfolio are equipped with the specified interfaces in order to be used with the DREAMing framework

Specification of interfaces for the planning functions and the DREAM-Bed

The specification of interfaces for the planning functions and the DREAMBed serves as the basis for the development of the planning DREAMBed and the interfacing of the planning methods

Specification of the interfaces for the navigation system and the DREAM-Bed

The specification of the interfaces for the navigation system and the DREAMBed is the first step in making the optimization framework accessible for navigation methods and viceversa

Evaluation report of the third stage of evaluations

The findings of the third stage of evaluations will be documented in this deliverable and will also be made available to the public

"Specification of the interfaces for the ""dreaming"" optimization system"

The specification of the interfaces for the dreaming optimization system is the basis is the basis for optimization algorithm development and the individual technology optimisations as well

Specification of perception interfaces

The interface specification for the perception methods is the basis for their integration with the DREAMing framework

Report on evaluation results

The overall results of the DREAMing evaluation of the perception methods will also be documented as the perception modality of the mobile manipulation system plays a special role compared to navigation and mobile manipulation

Mobile manipulation test board

The mobile manipulation test board is the first stage of the realistic manufacturing environment It present various typical mobile manipulation tasks in one mechanical setup which can also be replicated by the partners for individual evaluations

Realistic manufacturing environment

The realistic manufacturing environment RME replicates a typical setup from manufacturing situations where the mobile manipulator has to carry out tasks It will be set up for the project and will also be available at throughout the project runtime to the partners for continuous integration and evaluation

Final system for algorithm configuration including the improvements integrated by tests in case studies

The final system for algorithm configuration including the improvements integrated by tests in case studies has its roots in WP2 and comprises the final result the work on the algorithm configuration system for optimisation

Project material (templates, project flyer)

The project material logo templates and project flyer will be designed produced and distributed within the consortium for a clear presentation of the activities carried out within the project

Engineering framework for setting up, validating and optimising mobile manipulation tasks

The engineering framework for setting up validating and optimising mobile manipulation task will be developed and implemented by KUKA as an engineering access to the DREAMing framework and its capabilities

Realistic manufacturing environment design

The realistic manufacturing Environment RME is documented in this deliverable This documentation serves as a basis for the construction of the RME and the design of the evaluation use cases

Project website

The project website will help to disseminate the projects achievements and will also display the current project status A consortium internal part of the website will be utilised for organisational and collaboration purposes

Pubblicazioni

RoBO: A Flexible and Robust Bayesian Optimization Framework in Python

Autori: A. Klein and S. Falkner and N. Mansur and F. Hutter
Pubblicato in: NIPS 2017 Bayesian Optimization Workshop, 2017
Editore: NIPS

Unlocking the Potential of Simulators: Design with RL in Mind

Autori: Rika Antonova, Silvia Cruciani
Pubblicato in: Extended abstract for RLDM17 (3rd Multidisciplinary Conference on Reinforcement Learning and Decision Making), 2017
Editore: RLDM

Active classifier selection for RGB-D object categorization using a Markov random field ensemble method

Autori: Maximilian Durner, Zoltán Márton, Ulrich Hillenbrand, Haider Ali, Martin Kleinsteuber
Pubblicato in: Ninth International Conference on Machine Vision (ICMV 2016), 2017, Pagina/e 103411I
Editore: SPIE
DOI: 10.1117/12.2268551

Selecting CNN features for online learning of 3D objects

Autori: Monika Ullrich, Haider Ali, Maximilian Durner, Zoltan-Csaba Marton, Rudolph Triebel
Pubblicato in: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017, Pagina/e 5086-5091, ISBN 978-1-5386-2682-5
Editore: IEEE
DOI: 10.1109/IROS.2017.8206393

Pivoting Objects with a Parallel Gripper Using Controlled Slip and Inertial Forces

Autori: S. Cruciani and C. Smith
Pubblicato in: Extended Abstract in the IROS 2017 workshop: Frontiers in Contact-rich Robotic Interaction: Modeling, Optimization and Control Synthesis, 2017
Editore: IROS

Combining Hyperband and Bayesian Optimization

Autori: Stefan Falkner and Aaron Klein and Frank Hutter
Pubblicato in: Proceedings of the NIPS 2017 Bayesian Optimization Workshop, 2017
Editore: NIPS

Experience-based path planning for mobile robots exploiting user preferences

Autori: Lorenzo Nardi, Cyrill Stachniss
Pubblicato in: 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2016, Pagina/e 1170-1176, ISBN 978-1-5090-3762-9
Editore: IEEE
DOI: 10.1109/IROS.2016.7759197

Learning Curve Prediction with Bayesian Neural Networks

Autori: Aaron Klein, Stefan Falkner, Jost Tobias Springenberg & Frank Hutter
Pubblicato in: Proceedings of the International Conference on Learning Representations (ICLR) 2017, 2017
Editore: ICLR

Topometric Localization with Deep Learning

Autori: Gabriel Leivas Oliveira, N. Radwan, W. Burgard, Thomas Brox
Pubblicato in: Proceedings of the International Symposium on Robotics Research (ISRR) 2017, 2017
Editore: ISSR

Bayesian Optimization with Robust Bayesian Neural Networks

Autori: Springenberg, J.T., Klein, A., Falkner, S. and Hutter, F.
Pubblicato in: Advances in Neural Information Processing Systems 29 (NIPS 2016), 2016
Editore: eural Information Processing Systems Foundation, Inc.

Integrating motion and hierarchical fingertip grasp planning

Autori: Joshua A. Haustein, Kaiyu Hang, Danica Kragic
Pubblicato in: 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, Pagina/e 3439-3446, ISBN 978-1-5090-4633-1
Editore: IEEE
DOI: 10.1109/ICRA.2017.7989392

In-hand manipulation using three-stages open loop pivoting

Autori: Silvia Cruciani, Christian Smith
Pubblicato in: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017, Pagina/e 1244-1251, ISBN 978-1-5386-2682-5
Editore: IEEE
DOI: 10.1109/IROS.2017.8202299

Fast Bayesian Optimization of Machine Learning Hyperparameters on Large Datasets

Autori: Aaron Klein and Stefan Falkner and Simon Bartels and Philipp Hennig and Frank Hutter
Pubblicato in: Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, 2017, Pagina/e 528-536
Editore: PMLR

Experience-based optimization of robotic perception

Autori: Maximilian Durner, Simon Kriegel, Sebastian Riedel, Manuel Brucker, Zoltan-Csaba Marton, Ferenc Balint-Benczedi, Rudolph Triebel
Pubblicato in: 2017 18th International Conference on Advanced Robotics (ICAR), 2017, Pagina/e 32-39, ISBN 978-1-5386-3157-7
Editore: IEEE
DOI: 10.1109/ICAR.2017.8023493

Robot Navigation in Hand-Drawn Sketched Maps

Autori: Boniardi, F., B. Behzadian, W. Burgard, and G. Diego Tipaldi
Pubblicato in: Proceedings of the European Conference on Mobile Robots, 2015
Editore: IEEE

Maximum Likelihood Remission Calibration for Groups of Heterogeneous Laser Scanners

Autori: Steder, B., M. Ruhnke, R. Kümmerle, and W. Burgard
Pubblicato in: Proceedings of the IEEE International Conference on Robotics and Automation, 2015
Editore: IEEE

Relative Topometric Localization in Globally Inconsistent Maps

Autori: Mazuran, M., F. Boniardi, W. Burgard, and G. Diego Tipaldi
Pubblicato in: Proceedings of the International Symposium on Robotics Research, 2015
Editore: Springer

Accurate Localization with Respect to Moving Objects via Multiple-Body Registration

Autori: Röwekämper, J., B. Suger, W. Burgard, and G. Diego Tipaldi
Pubblicato in: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2015
Editore: IEEE

LexTOR: Lexicographic Teach Optimize and Repeat Based on User Preferences

Autori: Mazuran, M., C. Sprunk, W. Burgard, and G. Diego Tipaldi
Pubblicato in: Proceedings of the IEEE International Conference on Robotics and Automation, 2015
Editore: IEEE

Viewing Direction Classification: Application to View Planning

Autori: Riedel, S., Z-C. Marton, and S. Kriegel
Pubblicato in: Proceedings of the IEEE International Conference on Robotics and Automation, 2016
Editore: IEEE

Streaming Monte Carlo Pose Estimation for Autonomous Object Modeling

Autori: Rink, C., and S. Kriegel
Pubblicato in: Proceedings of the Conference on Computer and Robot Vision, 2016
Editore: IEEE

Autonomous Indoor Robot Navigation Using a Sketch Interface for Drawing Maps and Routes

Autori: Boniardi, F., A. Valada, W. Burgard, and G. Diego Tipaldi
Pubblicato in: Proceedings of the IEEE International Conference on Robotics and Automation, 2016
Editore: IEEE

Multi-view Orientation Estimation using Bingham Mixture Models

Autori: Riedel, S., Z-C. Marton, and S. Kriegel
Pubblicato in: Proceedings of the IEEE International Conference on Automation, Quality and Testing, Robotics, 2016
Editore: IEEE

On the Evolution of Fingertip Grasping Manifolds

Autori: Hang, K., Haustein, J.A. , Li, M., Billard, A., Smith, C. and Kragic, D.
Pubblicato in: Proceedings of the IEEE International Conference on Robotics and Automation, 2016
Editore: IEEE

User preferred behaviors for robot navigation exploiting previous experiences

Autori: Lorenzo Nardi, Cyrill Stachniss
Pubblicato in: Robotics and Autonomous Systems, Numero 97, 2017, Pagina/e 204-216, ISSN 0921-8890
Editore: Elsevier BV
DOI: 10.1016/j.robot.2017.08.014

Detecting Changes in the Environment Based on Full Posterior Distributions Over Real-Valued Grid Maps

Autori: Lukas Luft, Alexander Schaefer, Tobias Schubert, Wolfram Burgard
Pubblicato in: IEEE Robotics and Automation Letters, Numero 3/2, 2018, Pagina/e 1299-1305, ISSN 2377-3766
Editore: IEEE
DOI: 10.1109/LRA.2018.2797317

Improving object orientation estimates by considering multiple viewpoints

Autori: Zoltán Csaba Márton, Serkan Türker, Christian Rink, Manuel Brucker, Simon Kriegel, Tim Bodenmüller, Sebastian Riedel
Pubblicato in: Autonomous Robots, Numero 42/2, 2018, Pagina/e 423-442, ISSN 0929-5593
Editore: Kluwer Academic Publishers
DOI: 10.1007/s10514-017-9633-1

Toward fully autonomous mobile manipulation for industrial environments

Autori: Andreas Dömel, Simon Kriegel, Michael Kaßecker, Manuel Brucker, Tim Bodenmüller, Michael Suppa
Pubblicato in: International Journal of Advanced Robotic Systems, Numero 14/4, 2017, Pagina/e 172988141771858, ISSN 1729-8814
Editore: SAGE Publishing
DOI: 10.1177/1729881417718588

Design, Execution, and Postmortem Analysis of Prolonged Autonomous Robot Operations

Autori: Sebastian G. Brunner, Peter Lehner, Martin J. Schuster, Sebastian Riedel, Rico Belder, Daniel Leidner, Armin Wedler, Michael Beetz, Freek Stulp
Pubblicato in: IEEE Robotics and Automation Letters, Numero 3/2, 2018, Pagina/e 1056-1063, ISSN 2377-3766
Editore: IEEE
DOI: 10.1109/LRA.2018.2794580

Monte Carlo Registration and Its Application with Autonomous Robots

Autori: Christian Rink, Simon Kriegel, Daniel Seth, Maximilian Denninger, Zoltan-Csaba Marton, Tim Bodenmüller
Pubblicato in: Journal of Sensors, Numero 2016, 2016, Pagina/e 1-28, ISSN 1687-725X
Editore: Hindawi Publishing Corporation
DOI: 10.1155/2016/2546819

Reinforcement Learning for Pivoting Task

Autori: R Antonova, S Cruciani, C Smith, D Kragic
Pubblicato in: 2017
Editore: arvix

È in corso la ricerca di dati su OpenAIRE...

Si è verificato un errore durante la ricerca dei dati su OpenAIRE

Nessun risultato disponibile