Intelligent Robotic Inspection for Foundation Industry Optimisation (IRIFIO)
The Foundation Industries – a term used by the UK Research & Innovation programme for Metal, Glass, Ceramics, Paper, Cement, and bulk Chemical industries, produce 28m tonnes of material per year, and are worth £52bn annually to the UK economy. Around 75% of the materials in the environment today have been made by one of these six key industries.
All this activity results in the production of around 50 million tonnes of CO2 per year. That represents 10% of the total CO2 emitted by UK homes and businesses. If we are to meet our net zero carbon targets, seismic change is needed within these industries to reduce the amount of carbon they emit. New processes are required to keep these industries competitive internationally and to address sustainability issues. This can be done in part through managing resource and energy efficiency, streamlining processes to reduce waste, reduce energy consumption, and reduce costs.
I3DR won part of Innovate UK’s (part of UK Research and Innovation), £8 million funding to develop existing intelligent robotic systems to improve production process within the these energy intensive industries.
As project lead, I3DR was joined by Lucideon, a materials and process consultancy, and Glass Technology Services, experts in glass manufacturing supply chain, both global leaders in their fields. The aim of the project was to adapt I3DR’s technologies developed to detect and identify defect detection in metals and apply them to the Glass and Ceramics sectors.
Toughened glass is derived from standard float glass and is mostly used in commercial applications such as automotive and construction where temperature, load, and impact must all be considered. The process involves forcing a surface layer of glass at least 0.1 mm thick into compression by ion exchange of the sodium ions in the glass surface with potassium ions (which are 30% larger), achieved by soaking the glass in a potassium nitrate bath. However, the process can incur nickel sulphide inclusions within the glass and over time, these defects can cause spontaneous failure, the results of which can be catastrophic. Current inspection methods have resulted in lengthly, energy intensive, destructive testing methods which are energy-intensive, wasteful, and messy. The aim of this group’s efforts was to introduce in-process defect detection allowing for corrective methods to be introduced earlier into the process, reducing waste, cost, and energy use.
Phobos is I3DR’s high-resolution stereo camera. This is paired with a proprietary stereo matching algorithm to produce highly defined maps of any given topography. Machine learning technology combined with the stereo vision system is capable of identifying nickel suphide inclusions within glass without the requirement for visual / operator inspection and destructive testing methods, producing a quality assurance process that is reliable and repeatable.