Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Autonomous AI-enabled Industrial Sorting Pipeline for Advanced Textile Recycling

Yannis Spyridis, Vasileios Argyriou, Antonios Sarigiannidis, Panagiotis Radoglou, Panagiotis Sarigiannidis

TL;DR

The paper tackles the growing challenge of textile waste by proposing an autonomous textile analysis pipeline that leverages robotics, spectral imaging, AI-based classification, laser segmentation, and a Digital Twin (DT) within an Industry 4.0-inspired five-layer architecture. It evaluates multiple CNN architectures (EfficientNet, ResNest, and two custom networks) on a spectral-band dataset, finding that ResNest delivers the best classification performance (accuracy ≈ 0.586, precision ≈ 0.670, F1 ≈ 0.618), while others underperform and ROC-AUC remains modest. The Digital Twin enables predictive analysis, fault detection, and scenario testing, with documented parameter ranges for conveyors, cameras, and laser segmentation to simulate system dynamics. Collectively, the integrated framework aims to improve sorting accuracy, throughput, and sustainability, supporting a circular economy for textiles, though further data and model refinements are needed to reach optimal performance.

Abstract

The escalating volumes of textile waste globally necessitate innovative waste management solutions to mitigate the environmental impact and promote sustainability in the fashion industry. This paper addresses the inefficiencies of traditional textile sorting methods by introducing an autonomous textile analysis pipeline. Utilising robotics, spectral imaging, and AI-driven classification, our system enhances the accuracy, efficiency, and scalability of textile sorting processes, contributing to a more sustainable and circular approach to waste management. The integration of a Digital Twin system further allows critical evaluation of technical and economic feasibility, providing valuable insights into the sorting system's accuracy and reliability. The proposed framework, inspired by Industry 4.0 principles, comprises five interconnected layers facilitating seamless data exchange and coordination within the system. Preliminary results highlight the potential of our holistic approach to mitigate environmental impact and foster a positive shift towards recycling in the textile industry.

Autonomous AI-enabled Industrial Sorting Pipeline for Advanced Textile Recycling

TL;DR

The paper tackles the growing challenge of textile waste by proposing an autonomous textile analysis pipeline that leverages robotics, spectral imaging, AI-based classification, laser segmentation, and a Digital Twin (DT) within an Industry 4.0-inspired five-layer architecture. It evaluates multiple CNN architectures (EfficientNet, ResNest, and two custom networks) on a spectral-band dataset, finding that ResNest delivers the best classification performance (accuracy ≈ 0.586, precision ≈ 0.670, F1 ≈ 0.618), while others underperform and ROC-AUC remains modest. The Digital Twin enables predictive analysis, fault detection, and scenario testing, with documented parameter ranges for conveyors, cameras, and laser segmentation to simulate system dynamics. Collectively, the integrated framework aims to improve sorting accuracy, throughput, and sustainability, supporting a circular economy for textiles, though further data and model refinements are needed to reach optimal performance.

Abstract

The escalating volumes of textile waste globally necessitate innovative waste management solutions to mitigate the environmental impact and promote sustainability in the fashion industry. This paper addresses the inefficiencies of traditional textile sorting methods by introducing an autonomous textile analysis pipeline. Utilising robotics, spectral imaging, and AI-driven classification, our system enhances the accuracy, efficiency, and scalability of textile sorting processes, contributing to a more sustainable and circular approach to waste management. The integration of a Digital Twin system further allows critical evaluation of technical and economic feasibility, providing valuable insights into the sorting system's accuracy and reliability. The proposed framework, inspired by Industry 4.0 principles, comprises five interconnected layers facilitating seamless data exchange and coordination within the system. Preliminary results highlight the potential of our holistic approach to mitigate environmental impact and foster a positive shift towards recycling in the textile industry.
Paper Structure (15 sections, 3 equations, 6 figures, 4 tables)

This paper contains 15 sections, 3 equations, 6 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Textile Analysis
  4. Textile Analysis Datasets
  5. Autonomous Integrated Textile Analysis
  6. Methodology
  7. Architecture
  8. AI Classification
  9. Training dataset
  10. Convolutional neural networks
  11. Digital Twin System
  12. Evaluation
  13. Classification Results
  14. Digital Twin Results
  15. Conclusion

Figures (6)

  • Figure 1: Autonomous textile analysis pipeline architecture.
  • Figure 2: Dataset class distribution.
  • Figure 3: Digital Twin visualisation.
  • Figure 4: ResNest confusion matrix.
  • Figure 5: ROC-AUC curves for the EfficientNet and ResNest models.
  • ...and 1 more figures