r.assaf's picture




My Bio: 

Currently an Early Stage Researcher at the University of Salford, Working on Highly Adaptable, Autonomous and Flexible maintenance systems by embedding cognitive and cybernetic capabilities.

I received my Bachelor's Degree of Engineering in Mechatronics and Embedded Systems from ESTIA Institute of Technology - France, Then a Master of Science in Robotics and Embedded Systems from The University of Salford - UK.

His Masters Dissertation covers in detail the process of predictive modelling using non-linear multivariate regression. It shows the successful predictive system that was built, as well as its accurate predictions of website traffic. It also explains what goes into building a successful predictive model, especially showing the importance and impact of feature engineering, getting, transforming and modelling data into relevant features to improve the results and outcome of machine learning algorithms.

While pursuing my education, I have experienced and lived in many countries Including the UK, France, Japan and Spain. Also, I held a data scientist position in Paris as Expert on Optimisation Algorithms and Artificial Intelligence in the domain of Media and Advertising.

He has worked on both professional and University projects, including:

• Media Market Research and Implementation of Machine Learning Algorithms.

• Designing and developing a system to control robots using hand gestures.

• Designing and Developing an Autonomous Mobile Robot controlled by Artificial Intelligence.

• Programming ARM Microcontrollers to control and stabilize Cameras mounted on drones.

• Designing and Developing a Solar tracking system for Solar Panels.

What I do in Smart-E: 

ESR 11: Mathematical models for self-healing robot cells

As an early stage researcher in SMART-e, I work on the project of Mathematical modelling for self-healing robot cell, which is part of work package 2 Reconfigurable and Logistics Robotics. The projects aims to take a significant step towards the development of a system that will plan and execute maintenance with minimum human intervention ‘exactly’ when it needs to be, in a manner that gives the appearance of “self-healing”.
The research will be on the Mathematical modelling of imperfect maintenance along with inventory optimisation, also predicting faults and failures beforehand, so we can generate an optimized maintenance plan which lowers maintenance costs along with further reducing downtime for the factory.

Research interests: 

Machine Learning, investigating deep learning approaches

Learning from data

Statistical modelling

General Artificial intelligence



Assaf, Roy, et al. "Wear rate-state interaction modelling for a multi-component system: Models and an experimental platform." IFAC-PapersOnLine 49.28 (2016): 232-237.



Professor Samia Nefti-Meziani
Professor Philip Scarf

First Name: 


Last Name: 



Member for
5 years 11 months