Automated, Digitised and Decentralised Circular Waste Sorting

自动化、数字化和分散式循环垃圾分类

• 批准号: 48704
• 金额: $ 28.62万
• 依托单位: RECYCLEYE LTD
• 依托单位国家: 英国
• 项目类别: Study
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=48704
• 关键词: Automated; Digitised; Decentralised; Circular; Waste

Given all material has value, waste should not exist -- simply materials in the wrong place. Yet, 485m tonnes of material p/a is still not recycled in the UK but sent to landfill, exported abroad, or even worse ends up in the ocean. Currently, the recycling industry relies on large-scale material recovery facilities (MRF) which sort recyclables into usable material (purity = value). However, these require years of planning, large capital investment, numerous expensive machines and still use an agency workforce to manually sort materials. Smaller-scale facilities use manual labour entirely to sort waste as the capital cost for a minimal working MRF is too high.With this project, Recycleye will disrupt the waste management industry by creating a rapidly deployable, decentralized, scalable, digital & fully-automated sorting solution - mini material recovery facilities (mini-MRF). Recycleye has already developed a state-of-the-art computer vision system using recent advances in deep learning and AI capable of classifying items by material type and brand. The project's key objectives are to augment this vision-system with a robotic arm. A conveyor will carry waste towards the system where the vision unit will detect the different types of recyclables. Then, using a robotic arm the waste types will be sorted into respective piles providing a sorted/pure stream of recyclables that can then be re-injected into the UK economy/ sold to reprocessors.The project will focus on replicating uniquely human abilities - which machines have up until now not been able to match. The eye: The Recycleye vision system is able to, for example, understand that a piece of cardboard is pizza box (and hence likely food contaminated) even if it is torn, half-hidden by another item and covered in dirt. The arm: A mechanical arm is the most efficient way of removing target materials without having to rely on a material's physical characteristics (densities, ferrousity, form factor, etc). The project will build and test a fully operational mini-MRF ready for mass production.By developing the low-cost and data-driven sorting solution the recycling industry desperately needs, Recycleye is building the UK's future waste management infrastructure and accelerating the country's transition towards a circular economy.

鉴于所有材料都有价值，废物不应该存在-只是材料放错了地方。然而，在英国，每年仍有4.85亿吨的材料没有被回收，而是被送往垃圾填埋场，出口到国外，甚至更糟的是最终进入海洋。目前，回收行业依赖于大型材料回收设施（MRF），将可回收物分类为可用材料（纯度=价值）。然而，这些需要多年的规划，大量的资本投资，众多昂贵的机器，仍然使用代理劳动力手动分拣材料。小规模的设施完全使用手工劳动来分类废物，因为最小的工作MRF的资本成本太高。通过这个项目，Rebeccleye将通过创建一个快速部署，分散，可扩展，数字化和全自动化的分拣解决方案来颠覆废物管理行业-迷你材料回收设施（mini-MRF）。Sparkleye已经开发了一个最先进的计算机视觉系统，利用深度学习和人工智能的最新进展，能够按材料类型和品牌对物品进行分类。该项目的主要目标是用机械臂来增强这个视觉系统。传送带将把废物运送到系统中，视觉单元将检测不同类型的可回收物。然后，使用机械臂将废物分类到各自的堆中，提供分类/纯可回收物流，然后可以重新注入英国经济/出售给再处理者。该项目将专注于复制人类独特的能力-机器到目前为止无法匹配。眼睛：例如，Sparkleye视觉系统能够理解一块纸板是比萨饼盒（因此可能是食物污染），即使它被另一个物品撕裂，半隐藏并覆盖着污垢。手臂：机械臂是去除目标材料的最有效方式，而不必依赖于材料的物理特性（密度、铁性、形状因子等）。通过开发回收行业迫切需要的低成本和数据驱动的分拣解决方案，Reiscleye正在建设英国未来的废物管理基础设施，并加速该国向循环经济的过渡。