Computational Intelligence and Applications Research Group Research activities within the CIA group addresses a number of problem domains and real-world applications and can be sub-categorised into the following themes: Ambient Intelligent (AmI) - In environments based on ambient intelligence, it is intended to investigate prediction techniques using computational intelligence methods where the behaviours of users are predicted. This kind of environment is called a predictive ambient intelligence environment, which can be categorised as the new third generation of smart environments. The new emerging environment can learn from environmental changes as well as behavioural patterns of occupants. Predictive ambient intelligence environments collect data acquired from a sensor network. Collected data include a variety of attributes, such as the environmental changes and occupants' interactions with the environment. These data are used in a learning approach to make a predictive ambient intelligence environment that can predict the occupancy of different areas, as well as requirements and interests of occupants at different times. This predictive feature, for example, can improve the performance of energy saving approaches in a smart environment; in addition, it enhances the convenience of occupants as well as security and safety. Fuzzy Markup Language (FML) - use of fuzzy logic theory together with abstraction data tools, such as the eXtensible Markup Language (XML), for supporting the development of distributed fuzzy systems in computing environments composed by a collection of heterogeneous hardware. Through this new concept, the typical vision of fuzzy logic controllers is enhanced, moving it from a purely functional viewpoint (e.g. fuzzifier, defuzzifier, inference engine, etc.) to an abstract and interoperable vision based on the concept of labelled tree, a data structure defined by means of the graph theory and used by XML for its modelling aims. Fuzzy Markup Language (FML) is a novel computer language for defining fuzzy controllers independently from hardware constraints. Computational Optimisation - Focuses on the development and application of heuristic search and optimisation methods. These methods include genetic algorithms, simulated annealing, ant colony optimisation,Tabu search and hybrid methodologies known as Memetic Algorithms. Many scientific and engineering problems can be viewed as search or optimisation problems, where an optimum input vector for a given system has to be found in order to optimise the system response to that input vector. Often, auxiliary information about the system, such as its transfer function and derivatives, is not known, also various measures might be incomplete and distorted by noise. This makes such problems difficult to solve by traditional methods. In such cases, approaches based on computational optimisation techniques have been shown to be advantageous compared to classical approaches. Problems amenable to solution by heuristic search and computational optimisation techniques occur in all areas of science and engineering where an optimum design of a component or product, or optimum system input or response, is required.
Typical Implementation on AAL
Activities of Daily Living
Activities of Daily Living Classification
For more details, contact Dr Ahmad Lotfi (email@example.com)
Made with FlippingBook flipbook maker