Using AI to Improve Post-consumer Plastics Recycling for Sustainability in a Smart Environment
摘要
The optimization of plastic waste recycling represents a challenge for sustainable circular economy systems. This chapter explores the potential use of machine learning and robotics applications to improve the recycling rate in existing lightweight packaging sorting facilities in Germany. The research identifies key challenges in the recycling of post-consumer plastics, focusing on the application of AI, machine learning and computer vision. It highlights the importance of open data in creating an intelligent environment conducive to efficient recycling practices. The research also explores the selection of optimal sensors—including RGB (red, green, blue), near infrared (NIR) and RGB depth sensors—to improve the identification and classification of plastic materials. It also discusses the selection process for a suitable gripper mechanism, which is crucial for automating the sorting process, and the journey from laboratory prototyping to the realisation of production-ready systems.