Mauro Innocente

Dr Mauro S. Innocente is a Senior Lecturer in Aerospace Engineering and a member of the Research Institute for Future Transport and Cities (FTC) at Coventry University. He is also the founder and leader of the Autonomous Vehicles & Artificial Intelligence Laboratory (AVAILab).

He received his building & structural engineering degree from the National University of the Northeast, Argentina, in 2003; his master’s degree in numerical methods for calculus and design in engineering from the Polytechnic University of Catalonia, Spain, in 2007; his PhD degree in Particle Swarm Optimisation from Swansea University, UK, in 2011; and his Postgraduate Degree in Academic Practice in Higher Education from Coventry University, UK, in 2017.

He worked as an Engineering Assistant for a civil engineering company in Argentina from 2001 to 2003; as a Research Assistant in the Analysis & Advanced Materials for Structural Design group (AMADE) at the University of Gerona, Spain, in 2006; as a Research Assistant in the Civil & Computational Engineering Centre (C²EC) at Swansea University, UK, in 2010; as a Research Officer in the Advanced Sustainable Manufacturing Technologies (ASTUTE) project also at Swansea University from 2010 to 2014; as a Research Associate in the Institute of Energy at Cardiff University, UK, from 2014 to 2015; and joined Coventry University, UK, in 2015 as a Lecturer in Aerospace Engineering and a member of the Research Institute for Future Transport and Cities. He was promoted to Senior Lecturer and founded the AVAILab in 2018.

He currently teaches modules in Applied Mathematical Modelling & Artificial Intelligence within MSc courses in Aerospace and in Control Engineering.

He is the Principal Investigator (PI) of the 5th LRF-ICON PhD Studentship Award for a project entitled:
“Efficiency Enhancement of Class-A Foams by Means of Metal Oxide Nanoparticles for Self-Organising Swarms of Autonomous Unmanned Aerial Vehicles to Fight Wildfires” funded by the LRF International Consortium of Nanotechnologies (awarded in 2019, fEC: £50,000).
Co-Investigator (Co-I): Dr Evangelos Gkanas.

As Co-I, he was awarded the EPSRC Centre for Power Electronics Early-Career Researchers Grant for a project entitled “Optimisable system-level thermal models for power electronic converters” (awarded in 2015, total fEC: £55,000 - Coventry University fEC: £50,000) funded through EPSRC research grant EP/K035304/1 carried out in 2016 in collaboration with Dr Daniel J. Rogers (PI) at the University of Oxford.

His research is concerned with algorithm developments in Particle Swarm Optimisation, and with the development of soft and natural computing, mathematical modelling, control and optimisation tools within a comprehensive spectrum of applications such as fire-propagation modelling, thermal modelling, optimal design of power converters, structural analysis and design, swarm robotics applied to fire-fighting, autonomous AI-enabled UAVs for predictive maintenance in railway tracks, and optimal conjoint design of electrified powertrains and associated energy management strategies.

Engineering is a problem-solving discipline which applies scientific principles and entails a number of activities such as analysis, design, manufacturing and decision-making. Historically, the main concern was simply to attain the objective whilst improvement required iterative, time-consuming and costly processes. The ever-increasing demand to improve efficiency and lower costs has led engineers to seek rigorous design and decision-making methods. Thus, an aerospace engineer may seek the optimal aerofoil shape of an aircraft wing to generate the desired aerodynamic forces, or the lightest possible design of its inner structure capable of resisting the expected loads; an electrical engineer may seek the design of a power converter with the highest power-density possible while meeting required specifications; whereas an operational researcher may be interested in the most efficient configuration of their production lines. These kinds of problems can be dealt with through modelling, simulation and optimisation. Modelling offers a simplification of the system that keeps its most salient features; simulation uses the model to study its behaviour and performance by manipulating variables which would be impossible or impractical otherwise (descriptive); and optimisation techniques are used to identify the values of the variables that lead to a behaviour of the system that is best in some pre-defined sense (prescriptive). Thus, modelling, simulation and optimisation techniques are used in a wide range of industries such as civil, structural, aerospace, electrical, mechanical, automotive and manufacturing, to name a few.

While real-life problems typically do not lend themselves to elegant analytical models, the rapidly growing computing technology enables the use of mechanistic models with the aid of numerical methods such as the finite difference and finite element methods, and also the use of data-driven models such as polynomial response surfaces, Kriging models and Artificial Neural Networks (surrogate models). Increasing computing power also allows for the use of population-based optimisation methods, which are able to deal with previously intractable problems.

My fundamental research is aimed at algorithmic developments of the Particle Swarm Optimisation method towards a fully adaptive version that exploits its object-oriented nature and exhibits very good performance over a wide range of problems whilst relieving the end-user of the burden to problem-tune its control parameters. In turn, my applied research mainly targets the development of natural computing, mathematical modelling and optimisation tools within a comprehensive spectrum of applications spanning disciplines such as operational research, aerospace, structural, mechanical and electrical engineering. Examples are:

• Development of a fully adaptive general-purpose Particle Swarm Optimiser.
• Development of a fully adaptive surrogate-based optimal design tool for data-driven modelling and optimisation of problems whose mechanistic models are impractical or whose experimental data availability is limited.
• Development of simulation-based design tool for power-dense converters using efficient physics-based models.
• Mathematical modelling of the propagation of wildfires.
• Self-organising swarms of autonomous firefighting drones.
• Efficiency enhancement of class-A foams by means of nanoparticles to improve performance of firefighting drones.
• Autonomous AI-enabled UAVs for predictive maintenance in railway tracks.