One of the biggest challenges associated with creating effective material flows, where products and components are kept at their highest value for as long as possible cycling through the economy, is the separating out of different streams. Part of the solution is to make the flows effective to the point where a large mixture of end-of-life products is never created in the first place. However, it feels likely that waste material mixtures will always exist at some level. One potential opportunity, now being exploited by Eugenio Garnica from Sadako Technologies, is to utilise the potential of robotics and deep learning machines to sort through waste for high-value materials and components.
Sadako Technologies is currently working to re-purpose robots to monitor recycling conveyor belts and to recognise and pick out valuable materials and products. They claim to be able to cut the costs of sorting by more than 50%, changing the industry’s economics substantially.
The recycling machines require a level of artificial intelligence to recognise which items need to be retrieved.
The technology is only likely to improve and it’s an area of robotics that isn’t an obvious to be a threat to human-owned jobs. Sadako Technologies’ robots could just help to purify some of the more challenging mixed material flows.
If you’re interested in the subject of robots and machine learning, join a pair of DIF 2015 online headline act events on November 18th. At 15:00 GMT, Michelle Unger, the voice of Watson will talk about IBM’s remarkable machine, followed at 16:30 GMT by Jeremy Howard, CEO at Entilic and Rand Hindi, CEO at Snips, discussing how deep learning is changing lives.
Source: Circulate News RSS