​ ABOUT | SENSOR NETWORK | PUBLICATIONS
ABOUT
​
RAMSES4CE IN A NUTSHELL
​
The rapid identification of critical compounds is crucial for an adequate sorting and inherently the adapted recycling that will enable Circular Economy. RAMSES-4-CE is a 4 years up-scaling project funded by EIT RawMaterials to bring latest research into technical solutions for the resource industry. In this context, the RAMSES-4-CE project innovates optical spectroscopy-based multi-sensor systems for the recycling industry.​
​​​
​
WE FOCUS ON
​
1. developing a Raman sensor
2. Its integration in a LiF-HSI system (EIT inSPECtor)
3. advanced multi-source data fusion + machine learning for rapid data integration
​
​​​​​
​
BACKGROUND
​​
The steeply increasing demand for electronic devices and equipment, combined with the recent rise of high-technology greedy societal orientations like e-mobility and energy transition lead to vast amounts of e-wastes (WEEE). Most of the million tons of WEEE generated annually are only partially recycled up to now. This impedes dramatically EU goals towards a Circular Economy. In order to improve recycling efficiency and thus minimize our environmental footprint, modern recycling plants need multi-component sensors that can identify rapidly complex materials accurately.
​
Accordingly, new solutions for increased recycling rates need to overcome time- and cost-intensive existing material detection methods, which are incompatible with the urgent need to recycle increasingly complex industrial waste and automate adaption procedures when the composition changes.
​​
​
​OUR SCOPE
​
We address this challenge by using a combination of imaging sensors to identify key chemical compounds in material streams. By means of hyperspectral (HSI) reflectance spectroscopy in the near- and mid-infrared range certain alloys, ceramics, and plastics can be identified and classified. Laser-induced Fluorescence (LiF) spectroscopy enables the detection of rare earth elements (REEs) and low-reflective black plastics among others. In order to increase the range of waste classes characterized by our system, we propose to add a rapid, non-destructive and cost-efficient Raman sensor within the project RAMSES-4-CE. This module will be integrated into an existing sensor system, comprising LiF and HSI, developed during the EIT inSPECtor project.
​
For industrial applications, the requirements for a sensor-based sorting system implies high measurement speed (up to 1 m/s) for inline high throughput processing, as well a high spatial resolution (about 2 mm) for the identification of shredded and non-shredded recycling materials. Thus, the data generated by the individual sensors must be processed, integrated and analysed immediately. For this purpose, fast data processing tools based on machine learning will be developed. Together, smart real time integration of imaging and spectroscopic sensors using machine learning methods will allow for a robust analysis of complex, temporally and spatially variable industrial waste streams on a conveyor belt.
​
​
​
PROJECT PARTNERS
​​
​
​
​
​
​
​
​
​
FUNDED BY
​​​
​​​
​​​
​​
​​
​
​
​
​
​
​
​
_2023.jpg){kind=logo}
SENSOR NETWORK​
RAMSES4CE sensors
RAMSES is a fully integrated sensor system comprised of three optical characterisation technologies:
-
HSI: rapid identification of domains of interest and detection of certain alloys,ceramics, and plastics
-
LiF: detection of REEs and plastics
-
Raman: highly specific identification of organic (plastics) and inorganic compounds
​
​​​
​
​
​
​
​
​
​
​​
​
​
The smart sensor operates in sequential component mapping mode with immediate data fusion and result evaluation by machine learning methods, enabling:
-
rapid identification of domains for which a detailed chemical validation using Raman is required
-
integration of detailed chemical information to update domain characterization
-
creation of a dictionary learning of characteristic waste types
​​
​
​
​
​
​
​
​
​
​
​
​
​
ADVANTAGES of the RAMSES4CE SENSOR NETWORK
​​
-
Highly specific identification of materials using data integration to complement and cross-validate results
-
Smart multi-sensor approach delivers a flexible adaptive concept which allows for an efficient characterisation of several recycling waste streams
-
Sensor combination of HSI, LiF and Raman allows for simultaneous identification of organic and inorganic compounds
-
Machine learning methods allow for immediate identification of material mixtures
-
The non-invasive sensor network enables digitalization of material streams and hence, to pass information to subsequent processing routines and their adaption
-
Operation compatible with conveyor belt speeds
-
Sensors can be retrofitted to operate in existing sensor networks
​
​
Who can benefit from it?
​​
-
Recycling industries and geological exploration enterprises can benefit from the rapid and robust material identification achieved by the RAMAN smart sensor.
-
Industries aiming for rapid identification of alloys, ceramics, plastics, REEs and other organic and inorganic compounds. ​​​​​​​​​​​​​​​​​
​​
​​
​​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
PUBLICATIONS
​
​
2024
​
SCIENTIFIC PAPER - WASTE MANAGEMENT
A. de Lima Ribeiro, M. Fuchs, S. Lorenz, C. Röder, J. Heitmann, R. Gloaguen: Multi-sensor characterization for an improved identification of polymers in WEEE recycling. Waste Management, volume 178 (239–256), 2024. doi: 10.16/j.wasman.2024.02.024
​
BENCHMARK DATASET: PCB-Vision
Elias Arbash, Margret Fuchs, Behnood Rasti, Sandra Lorenz, Pedram Ghamisi, Richard Gloaguen: PCB-Vision: A Multiscene RGB-Hyperspectral Benchmark Dataset of Printed Circuit Boards. in IEEE Sensors Journal, doi: 10.1109/JSEN.2024.3380826
-
https://rodare.hzdr.de/record/2704 (dataset)
-
https://www.iexplo.space/post/towards-a-generalised-processing-of-hyperspectral-data-part-2 (blog)
​​
​
2023
​
WHISPERS 2023
Elias Arbash, Andréa de Lima Ribeiro, Sam Thiele, Nina Gnann, Behnood Rasti, Margret Fuchs, Pedram Ghamisi, Richard Gloaguen: Masking Hyperspectral Imaging Data with Pretrained Models. doi: 10.48550/arXiv.2311.03053
​
DATA: SPECTRAL LIBRARY - POLYMERS
A. de Lima Ribeiro, M. Fuchs, S. Lorenz, C. Röder, J. Heitmann, R. Gloaguen: Multi-sensor spectral database of WEEE polymers [Data set]. Rodare. http://doi.org/10.14278/rodare.2448
​
SMSI 2023
A. de Lima Ribeiro, M. Fuchs, S. Lorenz, R. Gloaguen, C. Röder, J. Heitmann, N. Schüler: How can Raman spectroscopy support optical detection systems for plastic identification in complex recycling streams? SMSI 2023, 2023-05-08 - 2023-05-11, Nürnberg, 217 - 218, doi: 10.5162/SMSI2023/D3.3
​
EGU 2023
Margret C. Fuchs, Sandra Lorenz, Yuleika C. Madriz Diaz, Titus Abend, Junaidh Shaik Fareedh, Andrea de Lima Ribeiro, Elias Arbash, Behnood Rasti, Jan Beyer, Christian Röder, Nadine Schüler, Kay Dornich, Johannes Heitmann, and Richard Gloaguen: How can agile sensing improve recycling stream characterisation and monitoring for e-waste? - news from the HELIOS lab. EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-12298
Andréa de Lima Ribeiro, Margret Fuchs, Christian Röder, Nadine Schüler, Sandra Lorenz, Yang Xiao Sheng, Johannes Heitmann, Kay Dornich, and Richard Gloaguen: ​Potential of optical sensors for polymer type identification in e-waste recycling streams. EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-12643
2022
​
WHISPERS 2022
Elias Arbash, Margret Fuchs, Behnood Rasti, Andrea de Lima Ribeiro, Pedram Ghamisi, Richard Gloaguen: Developing an RGB-Based PCB Objects Detector For Evaluation Analysis Guidance in a Smart
Sensors Network.
​
Andrea de Lima Ribeiro, Margret C. Fuchs, Sandra Lorenz, Yuleika Madriz, Erik Herrmann, Richard Gloaguen: Integrated hyperspectral and Raman sensors for fast characterization of plastics in e-waste recycling streams.
​
Margret C. Fuchs, Sandra Lorenz, Yuleika C. Madriz Dias, Titus Abend, Junaidh Shaik Fareedh, Andrea de Lima Ribeiro, Erik Herrmann, Elias Arbash, Seema Chouhan, Behnood Rasti, Jan Beyer, Christian Röder, Tejas Wakde, Nadine Schüler, Pedram Ghamisi, Kay Dornich, Johannes Heitmann, Richard Gloaguen: Smart hyperspectral sensor integration - insights from the HELIOS lab.
​
EGU 2022
Margret C. Fuchs, Andrea de Lima Ribeiro, Elias Arbash, Christian Röder, Nadine Schüler, Kay Dornich, Xiaosheng Yang, Richard Gloaguen, Johannes Heitmann: The RAMSES-4-CE project – developing a smart sensor network for e-waste characterisation. EGU22-13539, doi: 10.5194/egusphere-egu22-13539
​
SPIE.Photonics Europe 2022
Andréa de Lima Ribeiro, Margret C. Fuchs, Sandra Lorenz, Christian Röder, Yuleika C. Madriz, Erik Herrmann, Richard Gloaguen, Johannes Heitmann (2022): Multisensor characterization of WEEE polymers: spectral fingerprints for the recycling industry. Paper 12138-47
​
ACAMONTA 2022
Christian Röder, Margret Fuchs, Titus Abend, Jan Beyer, Nadine Schüler, Sandra Lorenz, Kay Dornich, Richard Gloaguen, Johannes Heitmann (2022): Kontaktlose Materialidentifikation und Digitalisierung für die Realisierung geschlossener Stoffkreisläufe. ACAMONTA 29, p. 23 - 26.