Dr Samir Khan

Dr Samir Khan

Associate Professor at the University of Tokyo

Portfolio

Application of deep learning in system health management

Published:

I am investigating various deep learning architectures including autoencoders and recurrent neural networks so that they can used for fault detection and isolation. This becomes important in the systems development life cycle if artificial intelligence is to be used to detect anomalies, analyse faults/failures and predict the remaining useful life of system components. By utilising historical data, data models are trained using machine learning and statistical concepts. Once trained, the logic for data processing can be embedded on on-board microcontrollers whilst enabling real-time health assessment and analysis. The work has a focus towards aerospace applications, for increasing overall system resilience and potential cost benefits for maintenance, repair and overhaul activities.

Collaborative autonomous robotics for maintenance

Published:

This project was about automating a maintenance tasking, with an investigation into methods for capturing and classifying maintenance processes and decomposing them into ‘unit tasks’ for future automation. This helps designers in linking these unit tasks with automation during conceptual design of autonomous robotic platforms. These concepts were demonstrated using multiple collaborative mobile manipulator robots (Turtlebots) to perform the simple industrial maintenance task using.

Non-uniform sampling strategies for digital control

Published:

Most conventional control algorithms cause numerical problems where data is collected at sampling rates that are substantially higher than the dynamics of the equivalent continuous-time operation that is being implemented. This is of relevant interest in applications of digital control, in which high sample rates are routinely dictated by the system stability requirements rather than the signal processing needs. Digital control systems exhibit bandwidth limitations enforced by their closed-loop frequency requirements that demand very high sample rates. Considerable recent progress in reducing sample frequency requirements has been made through the use of non-uniform sampling schemes, so called alias-free signal processing. The approach prompts the simplification of complex systems and consequently enhances the numerical conditioning of the implementation algorithms that otherwise would require very high uniform sample rates. However, the control communities have not yet investigated the use of intentional non-uniform sampling. The purpose of this article is to address some algorithmic issues when using such regimes for digital control.

Reduction of No Fault Found (NFF) through improved design

Published:

The research project addressed the inherent causes of NFF resulting from inadequacies in electronic equipment design. For example the design of a Built-In-Test-Equipment (BITE) which relies upon an in-depth knowledge of all system interactions. There also is the inevitable case where failure mechanisms only manifest themselves during a limited envelope of operation. These scenarios may not be recognised during design due to limits on system understanding resulting in the selection of inappropriate detection and confirmation strategies leading to a NFF susceptible design. Other areas which are of importance include inappropriate designed limit, software coding errors, inadequate procedures and when the process depends on human judgement for design and validation activities.