Multipurpose robotics for mAniPulation of defoRmable materIaLs in manufacturing processes

The APRIL project aims at prototyping low cost and agile market oriented multipurpose, and easy to repurpose, autonomous dexterous robots; which will manipulate, assemble or process different soft and flexible products/materials in a production line environment. This will enable new ways of automatization (semi- or fully-automatic tasks) in manufacturing lines that produce, assemble or handle different types of flexible or deformable materials (e.g., from pillows to delicate food products).

Duration: 01.04.2020 till 31.07.2023
Donee: German Research Center for Artificial Intelligence GmbH
Sponsor: European Union
Grant number: 870142

Universidad Politécnica de Madrid
Shadow Robot Company Ltd.
Prensilia SRL
Tree Technology SA
Fondazione Instituto Italiano di Technologia
Przemyslowy Instytut Automatyki i Pomiarow PIAP
Scuola Superiore di Studi Universitari e di Perfezionamento Sant’Anna
Instituto Tecnologico del Calzado y Conexas
Kontor 46 di Bonasso Matteo SAS
Silverline Endustri Ve Ticaret A.S.
Asociacion De Investigacion De Industrias Carnicas Del Principado De Asturias
Pemu Muanyagipari Zartkoruen Mukodoreszvenytarsasag
Osai Automation System SPA
Imprensa Nacional - Casa Da Moeda S.A.

Application Field: Logistics, Production and Consumer

Project details

To achieve this goal, APRIL will design and build robot prototypes optimized for processing soft and malleable materials in manufacturing environments; they will be validated and tested in the laboratory and in six different operational manufacturing environments and sectors (appliances, food, textiles, footwear, electronics and paper/passports) in five countries and in manufacturing industries of different sizes. These prototypes will be:

(1) be safe to use in the vicinity of unattended persons;

(2) be more skilled and competent in handling different types of soft products while controlling the degree of deformation;

(3) be better able to monitor any additional information of the product (colour, healthy condition of the food, etc.) during handling;

(4) be able to learn in new interactive environments; and

(5) be better able to move from line to line and from workstation to workstation as required



Grasping 3D Deformable Objects via Reinforcement Learning: A Benchmark and Evaluation
Melvin Laux, Chandandeep Singh, Alexander Fabisch
In 3rd Workshop on Representing and Manipulating Deformable Objects @ ICRA2023, (ICRA-2023), 29.5.-29.5.2023, London, ICRA, May/2023.


A Modular Approach to the Embodiment of Hand Motions from Human Demonstrations
Alexander Fabisch, Manuela Uliano, Dennis Marschner, Melvin Laux, Johannes Brust, Marco Controzzi
In Proceedings of the IEEE-RAS International Conference on Humanoid Robots, (Humanoids-2022), 28.11.-30.11.2022, Ginowan, IEEE, 2022. IEEE-RAS.


Sample-Efficient Policy Search with a Trajectory Autoencoder
Alexander Fabisch, Frank Kirchner
In Proceedings of the 4th Robot Learning Workshop: Self-Supervised and Lifelong Learning, (NeurIPS-2021), 14.12.2021, virtuell, n.n., Dec/2021.
gmr: Gaussian Mixture Regression
Alexander Fabisch
In Journal of Open Source Software, The Open Journal, Journal of Open Source Software, volume 6, number 62, pages 3054, Jun/2021.

Back to the list of projects
last updated 04.01.2024
to top