Deskilling robots in logistics environments
19 Jun 2020
Martin Davies, Senior Software Engineer at Guidance Automation recently contributed to The Scientific Journal ‘KI – Künstliche Intelligenz,’ – the official journal of the Division for Artificial Intelligence (AI) within the “Gesellschaft für Informatik e.V” – the German Informatics Society. The journal discusses all relevant aspects of AI, including use and adaption for scientific purposes and applications which are implemented using AI methods.
The Scientific Journal recently published a special issue on Reintegrating Artificial Intelligence and Robotics, which included contributions from a number of authors touching on the subject of robots, digital collaboration and AI. One of the editors of the special issue, Nick Hawkes, worked with Guidance Automation previously when we were trialling our automated vehicle solutions at Gompels Healthcare. Using our ‘Surface’ solution, vehicles were able to navigate around Gompels’ distribution centre using a floor code reading system via camera technology. The trial utilised new technologies to automate certain aspects of a typical warehouse operation in terms of the location, control, safe movement and supply of robotic vehicles.
Martin Davies contributed to the issue with his report on ‘Deskilling robots in logistics environments.’ His paper highlights the importance of deskilling the introduction and sustainability of robotic systems as a key success factor for replacing traditional static capital equipment. Robots are installed in logistics environments because they are adaptive; unlike their rigid mechanised counterparts such as conveyors and lifts. However, maintaining this adaptability post-installation and leaving it to the customer to reconfigure and maintain the system remains a difficult proposition. In conclusion, Davies focuses on how in order to be successful with automation, businesses must apply this within their planning and preparation stages.
Speaking to Martin about the motivation for the issue of his work, he commented: “I really wanted to get across the point that most organisations implement a mechanical static installation which remains in place for a number of years until it either breaks or becomes inefficient. The USP of robots is that they are dynamic and react to a changing work environment. However, the problem is that they require skilled operators to install and maintain them. Unless the customer is able to do this themselves, they don’t have a better solution than if they had installed static capital equipment.”
He carried on to say: “If the robot system is customer-friendly, getting contractors in to maintain static equipment or to maintain robots tends to the same thing. The robots have to be usable, like any other tool, for us to be able to sell large numbers and grow quickly as a company.”
To benefit from implementing robots in logistics environments, businesses need to have technical insight and visibility into errors which arise: “We can’t have robots that stop and say “error”, especially in warehouse settings. We need to know the reasons as to why a robot isn’t completing their task. Until you have a system that tells you where it is having trouble, then the customer’s response will always be phone support, with some sort of vague bug description. To benefit, you have to implement a system that supports itself so that we can work on product development.”
To read the journal in full, the publication is available at https://link.springer.com/journal/13218/33/4