The rapid introduction of digital technologies in manufacturing such as collaborative robotics and autonomous systems is exciting but also brings risks and uncertainties on cost, training and safety implications to the work place. The success of automation systems depends on many factors including the business case, human factors, process and product variability/complexity, to name a few. We aim to understand the complex inter-dependencies between various factors and translate the knowledge into novel tools and methods that can be adopted by businesses. The lab has extensive expertise in design process, life cycle costing, uncertainty analysis, systems modelling and knowledge management in High Value Manufacturing. We embrace digital technologies such as wearables, Virtual Reality and state-of-the-art data analytics and machine learning techniques to explore our understanding of complex processes and relationships in order to build quantifiable and robust models to support the design process.

Current PhD Students

Jordan Crawford
“A Systems Engineering Approach for Autonomous Technologies”

Distributed autonomous systems will become more common and pervasive in the future in many areas. The implementation of autonomous systems is complex due to the interactions between human, technical, social and environmental factors. There is a need to investigate these issues using a systems engineering approach and to ensure the design of the technologies are inherently safe, culturally acceptable in the workplace, and beneficial to the society. The thesis will investigate current issues related to autonomous technologies and propose a new framework to design, verify and validate these systems with human-in-the-loop.

Piotr Fratzcak
“Robot Learning with Human Emotions by Means of Immersive Virtual and Augmented Reality”

Ensuring safety and efficiency in human-robot collaboration (HRC) is essential to the future success of Robotic Autonomous Systems (RAS). This thesis exploits novel Immersive Virtual and Augmented Reality (IVAR) environments to assess and improve the suitability HRC in untested and potentially risky manufacturing tasks. The interdisciplinary research will leverage on low-cost IVAR environment and wearable hardware where virtual experiments will be designed to understand human responses in realistic industrial scenarios. This research will develop a functional HRC by enabling the robot to learn and adapt to human responses.

Simon Micheler 
“Early-Stage Decision Support for the Implementation of Intelligent Automation”

This research will contribute to the question of how manufacturing companies can identify business cases for intelligent automation. Currently, there is no systematic methodology to assess the critical influencing factors, and their interdependencies, that would need to be considered when implementing intelligent automation in the manufacturing context. A decision support tool for manufacturing businesses, which allows the user to identify business cases for intelligent automation as part of an early stage automation decision will be developed.

Past PhD Students

Angel Sanchez
“A Framework to Support Automation in Manufacturing through the Study of Process Variability”

A key challenge to manufacturing automation is the ability to adapt to process variability, which has relied on human operators’ skills to meet the product and process specifications. This thesis investigates the ‎variability involved in manual processes, and provides a framework to categorise and evaluate the process variability to suggest suitable level of automation.

‎Yuchen Zhao
“Human Skill Capturing and Modelling using Wearable Devices”

This thesis aims to reduce robot programming efforts significantly by developing a methodology to capture, model and transfer the manual manufacturing skills from a human demonstrator to the robot. Wearable sensors are investigated as a promising device to record the state-action examples without restricting the human experts during the skilled execution of their tasks. The thesis provides a methodology to produce a state-action model from skilled demonstrations that can be translated into robot kinematics and joint states for the robot to execute.