The desire for smart “lights out factories” which can autonomously produce components for high value manufacturing industries is described by the Industry 4.0 solution. This manufacturing methodology is appropriate for newly designed components, which take advantage of modern materials, robotic and automation processes, but not necessarily applicable to overhaul and repair. The aerospace overhaul and repair industry remains heavily dependent on human engineering skills to develop repair and re-manufacturing techniques for complex components of high value.
Development of any advanced, intelligent multi-agent robotic additive re-manufacturing system requires correct interrogation of metallic materials thermal properties, system control and output. Advanced programming of robots, data interpretation from associated sensory and feedback systems are required to mirror human input. Using process analysis to determine stimuli, replacement of human sensory receptors with electronic sensors, vision systems and high-speed data acquisition and control systems allows for the intelligent fine tuning of multiple heat input parameters to deposit the additive material at any one time. The interaction of these key components combined with novel robotic technology and experienced welding engineers has made possible the construction of a disruptive robotic re-manufacturing technology.
This paper demonstrates the design process and analyses the outputs sourced from observation and the recording of highly skilled human engineers when conducting manual remanufacturing and repair techniques. This data is then mined for the transferable control input parameters required to replicate and improve human performance.
This industry-academia research intensive collaboration between VBC Instrument Engineering Limited (UK) and The University of Sheffield has received project funding from the Engineering and Physical Sciences Research Council (EPSRC, 2006–2010), the Science and Facilities Technology Council (STFC, 2011–2013) and Innovate-UK with the Aerospace Technology Institute (2014–2018).
This paper can be accessed via the Springer website. [Paywall]
Presented at the AHFE 2018 conference.
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 793).