HarvestShield - AI-Powered Graphene Sensor for Early Detection of Berry Spoilage "FreshDetect"

点击翻译按钮获取中文摘要

Combating climate change globally is vital, and technology for better resource management can contribute to this effort. The recent COP26 report stresses the need to cut emissions by half within a decade and reach net-zero carbon emissions by 2050 to maintain global temperature rise below 1.5 degrees Celsius (1). Food waste plays a significant role in climate change, with 30% of global greenhouse gas emissions coming from food production, surpassing all commercial flights combined (2). Governments are pushing businesses to improve, with the UN's Sustainable Development Goal aiming to halve food waste and reduce food losses by 2030 (3). The problem resides in the supply chain, where 30-40% of food production is lost before reaching the market (4), leading to an annual loss of 1.6 billion tonnes of food worth $1.2 trillion (5).Altered Carbon's project focuses on developing the K9Sense chip, a groundbreaking digital nose technology utilising modified graphene and Machine Learning (ML) for gas-sensing applications across industries. The primary goal is to minimise fresh food waste in storage, significantly impacting the agricultural and fresh food supply chain sectors. The technology also has potential applications in fire detection, healthcare, and industrial monitoring.The K9Sense chip's innovation comes from its ability to detect various scents in real-time using a durable, solid-state multi-array sensor. Paired with advanced ML algorithms, the project aims to create a versatile, general-purpose sensor adaptable to multiple use cases, filling the gap in portable, intelligent, and customisable multi-element sensors for automation and industrial IoT applications.Altered Carbon will conduct lab experiments and case studies to test the chip and its associated hardware and software in controlled and real-world settings. This process will enable the company to enhance the K9 sensor chip's performance, refine its design, and develop a fully automated AI-based system for learning and identifying new scent triggers. The project also involves building a vast library of known scents and events, developing AI tools for scent detection and pattern analysis, and creating a cloud-based software system ready for commercial launch.