We all know that robots have come a long way - but how far?
Modern advances in technology are leading to highly complex systems that need to be controlled within rigorous specifications and under higher levels of autonomy. Such demands are bound to increase in future technology, making the design of automatic control systems ever more challenging. Some typical applications are robots that can move autonomously, which are being used to explore the surface of planet Mars or the control of prosthetic devices for disabled persons.
In reality, a controller that is able to meet such elaborate levels of performance and autonomy must exhibit features that are normally attributed to intelligence. These include the ability to adapt to unanticipated situations, to learn complex, previously unknown behaviour and the capacity to automatically plan reliable control strategies. In essence, such automatic controllers attempt to mimic the intelligent traits and self-organising features found in nature.
Simon Fabri believes that computational intelligence has developed as an interesting and promising processing tool for dealing with complex engineering systems.
Dr Ing. Fabri is leading a research project under the National RTDI Programme run by the Malta Council for Science and Technology (MCST) that focuses on the design and implementation of innovative techniques for computerised control of complex systems, using methodologies such as computational intelligence and stochastic control.
In this project, specific emphasis is being placed on artificial neural network techniques for learning reliable control strategies, aimed at guiding an autonomously mobile robot and for controlling the movement of a robotic arm, despite uncertainties or changes taking place in the physical structure of these robotic systems. A series of innovative control algorithms have been developed and tested by simulation. These are now being tested and evaluated on the robots.
Novel contributions from this research have been presented at several international conferences and seven papers have been published in peer-reviewed conference proceedings, with a number of others currently undergoing their review process. This research has the potential of transferring knowledge to local companies with the aim of upgrading their capacities for the design, development and implementation of novel and advanced automation and control products.
The team consists of Dr Ing. Fabri and Ing. Marvin Bugeja, also from the Department of Systems and Control Engineering at the University of Malta.
The Malta Council for Science and Technology is responsible for managing the local research funding programme, which has an annual budget of €700,000. This programme funds local research, preferably research that is collaborative in nature involving both academia and local industry. One of the successful projects funded through this programme is Computational Intelligence Techniques for Control of Complex Systems managed by Simon Fabri, head of the Department of Systems and Control Engineering at the University of Malta.
Computational intelligence, a computing methodology closely related to artificial intelligence, aims to develop machines that are able to learn and autonomously adapt themselves to changing environments.
Computational intelligence combines elements of learning, adaptation, evolution, autonomy and fuzzy logic to create programs that exhibit a degree of intelligence when handling complex situations. In other words such systems have, to some extent, the capacity to reason, plan, solve problems, etc.
For more details on the national R&I programme contact Eric Flask at the Malta Council for Science and Technology on tel. 2360 2123 or visit www.mcst.gov.mt.
Modern advances in technology are leading to highly complex systems that need to be controlled within rigorous specifications and under higher levels of autonomy. Such demands are bound to increase in future technology, making the design of automatic control systems ever more challenging. Some typical applications are robots that can move autonomously, which are being used to explore the surface of planet Mars or the control of prosthetic devices for disabled persons.
In reality, a controller that is able to meet such elaborate levels of performance and autonomy must exhibit features that are normally attributed to intelligence. These include the ability to adapt to unanticipated situations, to learn complex, previously unknown behaviour and the capacity to automatically plan reliable control strategies. In essence, such automatic controllers attempt to mimic the intelligent traits and self-organising features found in nature.
Simon Fabri believes that computational intelligence has developed as an interesting and promising processing tool for dealing with complex engineering systems.
Dr Ing. Fabri is leading a research project under the National RTDI Programme run by the Malta Council for Science and Technology (MCST) that focuses on the design and implementation of innovative techniques for computerised control of complex systems, using methodologies such as computational intelligence and stochastic control.
In this project, specific emphasis is being placed on artificial neural network techniques for learning reliable control strategies, aimed at guiding an autonomously mobile robot and for controlling the movement of a robotic arm, despite uncertainties or changes taking place in the physical structure of these robotic systems. A series of innovative control algorithms have been developed and tested by simulation. These are now being tested and evaluated on the robots.
Novel contributions from this research have been presented at several international conferences and seven papers have been published in peer-reviewed conference proceedings, with a number of others currently undergoing their review process. This research has the potential of transferring knowledge to local companies with the aim of upgrading their capacities for the design, development and implementation of novel and advanced automation and control products.
The team consists of Dr Ing. Fabri and Ing. Marvin Bugeja, also from the Department of Systems and Control Engineering at the University of Malta.
The Malta Council for Science and Technology is responsible for managing the local research funding programme, which has an annual budget of €700,000. This programme funds local research, preferably research that is collaborative in nature involving both academia and local industry. One of the successful projects funded through this programme is Computational Intelligence Techniques for Control of Complex Systems managed by Simon Fabri, head of the Department of Systems and Control Engineering at the University of Malta.
Computational intelligence, a computing methodology closely related to artificial intelligence, aims to develop machines that are able to learn and autonomously adapt themselves to changing environments.
Computational intelligence combines elements of learning, adaptation, evolution, autonomy and fuzzy logic to create programs that exhibit a degree of intelligence when handling complex situations. In other words such systems have, to some extent, the capacity to reason, plan, solve problems, etc.
For more details on the national R&I programme contact Eric Flask at the Malta Council for Science and Technology on tel. 2360 2123 or visit www.mcst.gov.mt.