Dr Giuliano Punzo

Department of Automatic Control and Systems Engineering

Lecturer

g.punzo@sheffield.ac.uk
+44 114 222 5233

Full contact details

Dr Giuliano Punzo
Department of Automatic Control and Systems Engineering
Amy Johnson Building
Portobello Street
Sheffield
S1 4DW
Profile

I am an aerospace engineer by training but, since I left the industry, I cultivated my interest in the broad area of control theory and complex systems. My current research aims at understanding complex dynamics at the interface between social and technical systems. These include the use of public services such as transportation or water systems, how these impact the demographic and the resilience of both the infrastructure and the community they serve. My research interests include network theory, control theory, consensus, complexity, swarm engineering and autonomous robotic systems.

Research interests
  • Complex Systems
  • Transportation and mobility
  • Dynamic Network and graph theory
  • Unmanned aerial systems
  • Distributed Control

Projects

The role of the road network on city economic performance

This project aims at mapping the transport enabled connectivity of the city economy. Using traffic data from Sheffield City Council and insights from the census, this project will draw a picture of mobility within Sheffield an across the City Region. The Network abstraction allows for the identification of critical nodes and links in the city fabric whereas the dynamics can be identified through the timeseries data. The project would deliver a modelling framework to complement and is some aspect advance the random walk assumption used to model the interaction between economic actors.

Decision making dynamics with non spatially embedded networks and superimposed information layers

Deciding between available alternatives, each more or less appealing than others depending on the proportion of people choosing it, is a classical decision and control problem.

However, in today's hyperconnected world, each decision maker is exposed to a wealth of easily available information, with either low reliability or filtered through cultural characteristics. Moreover, the individuals social networks are not spatially constrained, which means individuals capture information through the network from peers that potentially experience a completely different environment.

This project aims at producing a new understanding of the decision dynamics through a multilayer network approach where the perception of the option rewards is perceived and transmitted through direct exposure, through an imperfect information layer and a non spatially embedded social network. The preference for one choice over the others does not directly depends on the individuals experience of that option, nor solely on the experience of one node immediate neighbours. The project will relate the new insights in decision dynamics to the dynamics of information spreading, to the structure of the social network and the characteristic of the information layer.

Cooperative and non-cooperative approaches for the resilience of socio-technical systems

Users impose stresses on the technical systems and sustain the strain of the technical systems failure, when these happen. With the tragedy of the common at one extreme of non-cooperative exploitation of assets, and an exogenously optimised usage profile at the other extreme, there are many intermediate scenarios arising from the social dynamics amongst the users.

While cooperation is usually favourably regarded, it is very rarely established given the initial price in establishing it. What is the minimal intervention on the network, how and where should it be delivered to produce the onset of cooperation amongst player that preserves the health or enhance the performance of the technical systems they insist on?

Co-evolutionary dynamics in logistic networks

This project looks at developing innovative analytic methods to identify and map the end node resilience of logistic networks to the dynamics of intermediate nodes. The Northern corridor in East Africa is an archetype of logistic supply networks. Co-evolutionary dynamics of the logistic nodes and analysis of the network structure using algebraic graph theory are envisaged to be two of the possible tools to answer the research question.

Through the corridor, communities are ensured the continuity of supplies. Investments are attracted if a good level of resilience can be ensured for the logistic and supply chain. The sustained traffic volume increases the value of controlling the infrastructure assets.

This project is cast within the framework of evolutionary dynamics and complex network theory.
Evolutionary dynamics define state updating rules, that is, how the state of a node can change from zero to one as a function of the neighbour nodes’ states. Thus we can formally describe how a failure in one component affects the functioning of other neighbouring components. Such rules could also describe the probability that an alteration of the behaviour in one component compromises the functioning of a neighbouring component.

The fault propagation is linked to different properties of the network such as the node degree or the node centrality. These properties can be leveraged to measure the convergence probability to the all-mutant state in consequence of a mutant gene originating in a specific zone of the network. Irrespective of the application domain, the above property is a clear indication of the resilience of the network.

Publications

Journal articles

Conference proceedings papers

  • Clark R, Punzo G & Macdonald M (2016) Consensus speed optimisation with finite leadership perturbation in k-nearest neighbour networks. 2016 IEEE 55th Conference on Decision and Control (CDC), 12 December 2016 - 14 December 2016. RIS download Bibtex download
  • Clark R, Punzo G, Baumanis K & Macdonald M (2016) Consensus speed maximisation in engineered swarms with autocratic leaders. Proceedings of the International Conference on Artificial Intelligence and Robotics and the International Conference on Automation, Control and Robotics Engineering - ICAIR-CACRE '16, 13 July 2016 - 15 July 2016. RIS download Bibtex download
  • Clark RA, Punzo G, Dobie G, MacLeod CN, Summan R, Pierce G, Macdonald M & Bolton G (2015) 3D model generation using an airborne swarm. 41ST ANNUAL REVIEW OF PROGRESS IN QUANTITATIVE NONDESTRUCTIVE EVALUATION: Volume 34 RIS download Bibtex download
  • Karagiannakis P, Weiss S, Punzo G, Macdonald M, Bowman J & Stewart RW (2013) Impact of a Purina fractal array geometry on beamforming performance and complexity. European Signal Processing Conference RIS download Bibtex download
  • Pierce SG, Punzo G, Dobie G, Summan R, Macleod CN, McInnes C, Biggs J, Macdonald M & Bennet D (2012) Reconfigurable robotic platforms for structural health monitoring. Proceedings of the 6th European Workshop - Structural Health Monitoring 2012, EWSHM 2012, Vol. 2 (pp 1187-1194) RIS download Bibtex download
  • Punzo G, Dobie G, Bennet DJ, Jamieson. J & Macdonald M (2012) Low-cost, multi-agent systems for planetary surface exploration. Proceedings of the International Astronautical Congress, IAC, Vol. 11 (pp 9244-9255) RIS download Bibtex download
  • Punzo G (2011) Fractal patterns in fractionated spacecraft. 62nd International Astronautical Congress 2011, IAC 2011, Vol. 11 (pp 9034-9043) RIS download Bibtex download
  • Punzo G, Bennet DJ & Macdonald M (2011) A fractally fractionated spacecraft. 62nd International Astronautical Congress 2011, IAC 2011, Vol. 9 (pp 7147-7156) RIS download Bibtex download
  • Punzo G, Bennet D & Macdonald M (2011) Enhancing Self-similar Patterns by Asymmetric Artificial Potential Functions in Partially Connected Swarms (pp 390-391) RIS download Bibtex download
  • Punzo , Macdonald M & Bennet DJ (2010) swarm shape manipulation through connection control. TAROS 2010 RIS download Bibtex download
  • Lenguito G, Punzo G, Rega RC & Semeraro O (2004) Possible new developments of MPD propulsion. International Astronautical Federation - 55th International Astronautical Congress 2004, Vol. 13 (pp 8935-8941) RIS download Bibtex download
Teaching activities
  • AER120 Introduction to Aerospace Design
  • ACS230 Control Systems Design and Analysis
  • First, second and year abroad tutor for Aerospace Engineering